RTMPose
OpenMMLab's RTMPose (t, s, m, l; 256x192 and 384x288): real-time top-down 2D human pose, a CSPNeXt CNN with a SimCC head; also the whole-body DWPose and RTMW checkpoints (133 keypoints with feet, face and hands); Halpe (26 keypoints), hand, face and animal checkpoints. Apache-2.0.
stable Fork of open-mmlab/mmpose@5408bc7 · backends: portable, kernels
- Paper RTMPose: Real-Time Multi-Person Pose Estimation based on MMPose arXiv · first released 13 Mar 2023
Paper 2023 Added in 0.2.0 Evaluated with 0.2.0
Use it
pip install fastforks Lean install with only what this model needs: pip install fastforks-core
from fastforks import pose
model = pose.load("rtmpose-t") # kernels on a supported GPU, else portable PyTorch
result = model(image, boxes=...)
result.keypoints # also: keypoint_scores, boxes, scores
result.backend # what ran: mode, device, dtype, kernels fastforks bench rtmpose-t Times the model on this machine. load(…, mode="portable") or mode="kernels" picks a backend; usage guide.
Speed and memory
Every backend on one GPU (RTX 5060 Ti 16 GB) and the same input, the authors’ code included. Kernels runs fp16. Portable runs fp16 on CUDA.
Input: synthetic, 1080x1920 (the task’s standard size)
| Backend | dtype | Time per call (ms) | vs original fp16 | Peak memory allocated / reserved | Load (s) | First call (s) |
|---|---|---|---|---|---|---|
| Original fp32mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed call; ≈ 161 ms of it CPU crop + decode | fp32 | 165 | 0.91× | 475 MiB/524 MiB | 0.1 | 0.17 |
| Original fp16torch.autocast(fp16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed call; ≈ 146 ms of it CPU crop + decode | fp16 | 151 | – | 459 MiB/490 MiB | 0.1 | 0.15 |
| Original fp32, network onlycrop and decode not timed | fp32 | 4.02 | 38× | 118 MiB/150 MiB | – | 0.45 |
| Original fp16, network onlycrop and decode not timed | fp16 | 4.86 | 31× | 89 MiB/114 MiB | – | 0.80 |
| Portable fp32 | fp32 | not measured | – | – | – | – |
| PortableGPU crop 1.22 · network 2.91 · decode + map back 0.436 ms | fp16 | 4.60 | 33× | 100 MiB/128 MiB | 0.2 | 0.86 |
| KernelsGPU crop 1.24 · network + decode 1.31 · map back 0.325 ms | fp16 | 2.90 | 52× | 105 MiB/156 MiB | 0.7 | 0.29 |
| Portable + torch.compile | 1.91 | 79× | 125 MiB/128 MiB | 11 (build) | – | |
| TensorRT fp16network on person crops only | fp16 | 0.866 | 174× | 56 MiBenginenot comparable | 16 (build) | – |
| TensorRT + our crop and decodeestimate: engine 0.866 + portable GPU crop 1.22 + decode 0.436 ms | fp16 | 2.52 | 60× | – | – | – |
Measured on RTX 5060 Ti 16 GB, torch 2.14.1+cu130, 2026-11-10 to 2026-11-16. First call, 3 warm-up, peak reset, median of 3 rounds x median of 5. Ratios divide two
times of this table (same GPU, same input). Original fp16 is the authors’ own precision option, else torch.autocast. “Network only” times the original’s network on the same crops, batched as one call; the
original rows above also include its own crop pipeline and decode, as ours do. The splits under Portable and Kernels come from a separate run that times each step on its own
(synchronised in between); the TensorRT engine covers the network only, so its full call is an estimate with
our crop and decode added. torch.compile and TensorRT: build time instead of load time. TensorRT memory (engine, not comparable) is the GPU memory the engine took, read from the driver because TensorRT allocates outside PyTorch, so it can't be compared with the PyTorch peaks of the other rows.
Other devices
| Device | Backend | dtype | ms / call | Peak memory | Output vs reference | Input |
|---|---|---|---|---|---|---|
| Apple M1 Pro 16 GB MPS, torch 2.14.1, 2026-11-13 | Original | fp32 | 157 | 1,804 MiB process | – | 1080x1920 |
| Portable | fp16 | 25.6 | 1,460 MiB process | matches | 1080x1920 | |
| Portable | fp32 | 25.7 | 1,462 MiB process | matches | 1080x1920 | |
| Kernels | fp32 | 20.1 | 1,532 MiB process | matches | 1080x1920 |
CUDA memory is torch’s peak allocated / reserved; Apple has no such counter, so the Mac rows show the process footprint (weights, Python and torch included). “Output vs reference” compares the device’s output with the stored golden output under the card’s tolerance; – = not checked on that device. Kernels “not implemented” = no kernels for this model on that backend yet: it runs portable there.
Licence and weights
Commercial use allowed
- Code
- Apache-2.0 LICENSE
- Weights
- Apache-2.0 source
- Training data
-
- COCO 2017 (train, instance annotations) CC-BY-4.0
- AI Challenger human keypoints (2017) not stated
- CrowdPose not stated
- MPII Human Pose BSD-2-Clause, non-commercial
- sub-JHMDB (J-HMDB) not stated
- Halpe-FullBody not stated
- PoseTrack 2018 not stated
- ImageNet-1k (ILSVRC 2012 classification) noncommercial, non-commercial
Our reading of the licences, not legal advice.
Weights
rtmpose-t-body7-256x192.pth 13.5 MB· hosted on OpenMMLabfastforks downloads and checks these files on first use (fastforks download rtmpose-t).
Accuracy
The test set through the authors’ code and through ours: same checkpoint, same protocol. The distance between the original’s fp32 and fp16 runs is what half precision alone costs; our runs should stay within it.
COCO val2017
ground-truth person boxes, keypoint AP
| Run | dtype | keypoint AP ↑ | Δ keypoint AP vs fp32 original | images | ms / image |
|---|---|---|---|---|---|
| Original fp32 | fp32 | 65.2 | reference | 5000 | – |
| Original fp16 | fp16 | 65.1 | −0.1 | 5000 | – |
| Portable fp32 | not measured | ||||
| Portable | fp16 | 65.2 | ±0.0 | 5000 | 4.56 |
| Kernels | fp16 | 65.2 | +0.0 | 5000 | 2.08 |
Top-down on the ground-truth person boxes (non-crowd, at least one labelled keypoint), no flip test; people ranked by the mean of their 17 keypoint scores (not mmpose's rescoring and OKS-NMS), so not the papers' AP; pycocotools OKS AP; whole-body models scored on their 17 body keypoints. Original = the authors’ code, same checkpoint and protocol. Measured on original fp32, original fp16: RTX 5060 Ti, torch 2.14.1+cu130, 2026-11-11; portable, kernels: RTX 5060 Ti, torch 2.14.1+cu130, fastforks 2241b095+unc, 2026-11-09.
Closeness to the original
Beyond the test-set scores: the raw output of ours against the original’s fp32 output on the same input, pixel by pixel. CPU, float32: max |diff| <= 0.002 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.66 and p99 <= 3.6 (absolute error of the output keypoints).
| Check | Tolerance | Measured |
|---|---|---|
| CPU fp32, max |diff| | ≤ 0.002 | not recorded |
| GPU default dtypes, mean |diff| | ≤ 0.66 | 0.40835 |
| GPU default dtypes, p99 |diff| | ≤ 3.6 | 2.02585 |
| gpu_min_score | ≤ 0.3 | not recorded |
Recorded for rtmpose-t on RTX 5060 Ti, torch 2.14.1+cu130, 2026-11-11, with these kernels: cudnn_haccfvm_convfvm_extrafvm_mbfvm_r3
Details
RTMPose (Jiang et al. 2023): real-time top-down 2D human pose from OpenMMLab: a CSPNeXt CNN backbone and a SimCC head (a gated attention unit over one token per keypoint, then x and y classified as 1D bins at half-pixel resolution, no heatmaps). The body7 checkpoints (trained on seven public pose datasets) return the 17 COCO keypoints; the DWPose ones (Yang et al. 2023, distilled on COCO-WholeBody + UBody, the pose model of ControlNet's openpose preprocessors) the 133 COCO-WholeBody keypoints (body, feet, face, both hands). The RTMW ones (Jiang et al. 2024: a PAFPN neck and a two-scale head on the same backbone, trained on 14 datasets) return the same 133 keypoints more accurately. More model-zoo checkpoints of the same network: `-halpe26` (body + feet, 26 keypoints), `-hand5` (21 keypoints of a hand box), `-face6` (106 face landmarks of a face box) and `-ap10k` (17 keypoints of an animal box). Takes person boxes (e.g. from `fastforks.detection`) and crops each at 256 x 192 (or 384 x 288; hand, face and animal boxes: 256 x 256); vendored from mmpose (no mmpose / mmcv dependency). The portable `module()` maps crops to SimCC logits.
- Input / output
- Input: one RGB image and a box per person. Output: keypoints with confidence scores for every person.
- Input size
- any H x W >= 32, padded to a multiple of 32; tested up to 3000x4000 (453 MiB allocated)
- Options
flip_test=false· changes the output
- Outputs
keypoints,keypoint_scores,boxes,scores- Checkpoint
deepen=0.167,widen=0.375- CUDA kernels
cudnn_haccconvolutions with fp16 accumulation (cuDNN)fvm_convfused convolutions for image classifiersfvm_extrafused transformer row operations (rotary split, SwiGLU with LayerNorm)fvm_mbfused MBConv blocks (MobileNet and EfficientNet families)fvm_r3further fused classifier blocks
- Apple (MPS)
pose_metalMetal: top-down pose decoders and crops. fp32, fp16
Full reference, generated from the repository
OpenMMLab's RTMPose (t, s, m, l; 256x192 and 384x288): real-time top-down 2D human pose, a CSPNeXt CNN with a SimCC head; also the whole-body DWPose and RTMW checkpoints (133 keypoints with feet, face and hands); Halpe (26 keypoints), hand, face and animal checkpoints. Apache-2.0.
Commercial use: yes.
| variant | input | backend | time (ms) | peak memory (MiB, allocated / reserved) |
|---|---|---|---|---|
rtmpose-t | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 165 151 4.60 2.90 1.91 0.866 (network on person crops only) | 475 / 524 459 / 490 100 / 128 105 / 156 125 / 128 127 / 164 |
rtmpose-s | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 161 165 4.95 3.08 2.07 1.06 (network on person crops only) | 461 / 510 449 / 480 118 / 152 109 / 158 129 / 150 137 / 170 |
rtmpose-m | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 179 180 6.36 4.50 3.42 2.02 (network on person crops only) | 323 / 376 316 / 364 145 / 192 125 / 200 147 / 172 157 / 176 |
rtmpose-l | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 161 176 9.04 5.70 5.32 3.20 (network on person crops only) | 1076 / 1114 1105 / 1154 200 / 256 152 / 244 185 / 194 230 / 266 |
rtmpose-m-384 | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 174 167 13.3 9.23 7.09 4.33 (network on person crops only) | 585 / 666 505 / 582 224 / 332 178 / 320 190 / 212 212 / 242 |
rtmpose-l-384 | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 192 171 19.4 11.4 10.7 7.08 (network on person crops only) | 1485 / 1538 366 / 410 314 / 416 217 / 390 242 / 274 297 / 322 |
dwpose-t | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 160 174 4.60 3.17 1.96 0.940 (network on person crops only) | 841 / 874 832 / 858 104 / 132 109 / 174 129 / 148 136 / 160 |
dwpose-s | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 173 180 5.16 3.35 2.27 1.17 (network on person crops only) | 807 / 860 806 / 848 124 / 154 115 / 178 135 / 150 149 / 170 |
dwpose-m | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 169 164 7.31 4.86 3.80 2.13 (network on person crops only) | 746 / 802 756 / 792 153 / 196 133 / 200 156 / 174 203 / 228 |
dwpose-l | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 178 184 10.3 6.20 5.75 3.33 (network on person crops only) | 341 / 388 393 / 428 211 / 260 163 / 266 197 / 216 277 / 294 |
dwpose-l-384 | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 194 203 21.4 12.7 11.9 7.46 (network on person crops only) | 796 / 832 773 / 816 325 / 418 229 / 404 253 / 296 347 / 370 |
rtmw-m | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 169 205 10.9 7.54 6.23 3.34 (network on person crops only) | 288 / 328 379 / 410 181 / 244 164 / 266 183 / 202 231 / 256 |
rtmw-l | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 213 203 15.3 9.99 9.34 5.17 (network on person crops only) | 679 / 708 865 / 908 256 / 308 217 / 312 242 / 264 354 / 396 |
rtmw-x | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 218 207 24.0 14.4 15.3 8.47 (network on person crops only) | 1307 / 1360 604 / 636 335 / 390 289 / 392 320 / 328 497 / 514 |
rtmw-l-384 | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 274 275 31.8 20.6 19.5 11.2 (network on person crops only) | 982 / 1018 1094 / 1144 371 / 468 276 / 458 298 / 344 424 / 452 |
rtmw-x-384 | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 306 291 50.6 50.9 34.0 19.3 (network on person crops only) | 1705 / 1812 1874 / 1940 466 / 632 496 / 632 361 / 388 581 / 624 |
rtmpose-t-halpe26 | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 181 172 4.61 2.90 1.94 0.871 (network on person crops only) | 119 / 158 103 / 144 100 / 130 105 / 156 125 / 128 128 / 164 |
rtmpose-s-halpe26 | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 172 173 4.99 3.05 2.11 1.07 (network on person crops only) | 174 / 212 162 / 200 119 / 152 110 / 178 130 / 150 138 / 170 |
rtmpose-m-halpe26 | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 179 181 6.42 4.53 3.44 2.02 (network on person crops only) | 288 / 328 283 / 318 146 / 192 125 / 180 148 / 172 160 / 180 |
rtmpose-l-halpe26 | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 180 164 9.10 5.69 5.34 3.20 (network on person crops only) | 498 / 532 529 / 572 201 / 256 152 / 244 186 / 194 231 / 266 |
rtmpose-m-halpe26-384 | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 192 193 13.4 9.28 7.12 4.34 (network on person crops only) | 769 / 850 690 / 764 225 / 332 179 / 320 190 / 212 212 / 238 |
rtmpose-l-halpe26-384 | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 206 206 19.5 11.5 10.6 7.08 (network on person crops only) | 1019 / 1054 975 / 1038 315 / 416 218 / 410 243 / 276 299 / 322 |
rtmpose-x-halpe26-384 | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 247 217 33.8 19.1 21.1 13.3 (network on person crops only) | 1470 / 1588 470 / 556 385 / 478 280 / 448 262 / 298 430 / 456 |
rtmpose-m-hand5 | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 192 195 8.05 5.46 4.51 2.58 (network on person crops only) | 451 / 504 425 / 472 166 / 214 139 / 200 176 / 192 174 / 192 |
rtmpose-t-face6 | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 188 204 4.53 3.61 2.23 1.00 (network on person crops only) | 444 / 498 424 / 464 116 / 152 123 / 180 159 / 168 145 / 168 |
rtmpose-s-face6 | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 258 237 5.04 3.74 2.69 1.28 (network on person crops only) | 519 / 572 504 / 540 140 / 174 128 / 202 164 / 190 158 / 178 |
rtmpose-m-face6 | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 198 207 8.75 5.77 4.79 2.67 (network on person crops only) | 666 / 728 652 / 698 173 / 216 145 / 222 182 / 192 216 / 240 |
rtmpose-m-ap10k | synthetic (1080x1920) | original (fp32) original portable kernels portable + torch.compile TensorRT fp16 | 196 186 8.02 5.42 4.54 2.57 (network on person crops only) | 783 / 846 757 / 814 166 / 214 139 / 200 176 / 192 172 / 188 |
Details
- Original code: https://github.com/open-mmlab/mmpose at
5408bc76f5b8 - Code licence: Apache-2.0 (permissive); files:
LICENSE - Weights licence: Apache-2.0
- Licence evidence: https://github.com/open-mmlab/mmpose/blob/5408bc76f5b848cf925a0d1857899011d8c5b497/README.md#license
- Training data: COCO 2017 (train, instance annotations), AI Challenger human keypoints (2017), CrowdPose, MPII Human Pose, sub-JHMDB (J-HMDB), Halpe-FullBody, PoseTrack 2018, ImageNet-1k (ILSVRC 2012 classification) (body7: COCO, AI Challenger, CrowdPose, MPII, sub-JHMDB, Halpe and PoseTrack18 keypoints; the CSPNeXt backbone was pre-trained on ImageNet-1k (and on the same pose data,
pt-body7))dwpose-t: COCO-WholeBody, UBody (upper-body whole-body keypoints, from OSX), COCO 2017 (train, instance annotations), AI Challenger human keypoints (2017), ImageNet-1k (ILSVRC 2012 classification) (COCO-WholeBody and UBody whole-body keypoints, distilled from a larger RTMPose (DWPose two-stage distillation); the CSPNeXt backbone was pre-trained on ImageNet-1k and on AI Challenger + COCO body keypoints)dwpose-s: COCO-WholeBody, UBody (upper-body whole-body keypoints, from OSX), COCO 2017 (train, instance annotations), AI Challenger human keypoints (2017), ImageNet-1k (ILSVRC 2012 classification) (COCO-WholeBody and UBody whole-body keypoints, distilled from a larger RTMPose (DWPose two-stage distillation); the CSPNeXt backbone was pre-trained on ImageNet-1k and on AI Challenger + COCO body keypoints)dwpose-m: COCO-WholeBody, UBody (upper-body whole-body keypoints, from OSX), COCO 2017 (train, instance annotations), AI Challenger human keypoints (2017), ImageNet-1k (ILSVRC 2012 classification) (COCO-WholeBody and UBody whole-body keypoints, distilled from a larger RTMPose (DWPose two-stage distillation); the CSPNeXt backbone was pre-trained on ImageNet-1k and on AI Challenger + COCO body keypoints)dwpose-l: COCO-WholeBody, UBody (upper-body whole-body keypoints, from OSX), COCO 2017 (train, instance annotations), AI Challenger human keypoints (2017), ImageNet-1k (ILSVRC 2012 classification) (COCO-WholeBody and UBody whole-body keypoints, distilled from a larger RTMPose (DWPose two-stage distillation); the CSPNeXt backbone was pre-trained on ImageNet-1k and on AI Challenger + COCO body keypoints)dwpose-l-384: COCO-WholeBody, UBody (upper-body whole-body keypoints, from OSX), COCO 2017 (train, instance annotations), AI Challenger human keypoints (2017), ImageNet-1k (ILSVRC 2012 classification) (COCO-WholeBody and UBody whole-body keypoints, distilled from a larger RTMPose (DWPose two-stage distillation); the CSPNeXt backbone was pre-trained on ImageNet-1k and on AI Challenger + COCO body keypoints)rtmw-m: COCO-WholeBody, UBody (upper-body whole-body keypoints, from OSX), AI Challenger human keypoints (2017), CrowdPose, MPII Human Pose, sub-JHMDB (J-HMDB), Halpe-FullBody, PoseTrack 2018, COCO 2017 (train, instance annotations), ImageNet-1k (ILSVRC 2012 classification) (Cocktail14: AI Challenger, CrowdPose, MPII, sub-JHMDB, Halpe, PoseTrack18, COCO-WholeBody, UBody, Human-Art, WFLW, 300W, COFW, LaPa and InterHand2.6M (the last six are not in the dataset list; several are research-only), distilled from RTMW-l; the CSPNeXt backbone was pre-trained on ImageNet-1k and on body / whole-body keypoints)rtmw-l: COCO-WholeBody, UBody (upper-body whole-body keypoints, from OSX), AI Challenger human keypoints (2017), CrowdPose, MPII Human Pose, sub-JHMDB (J-HMDB), Halpe-FullBody, PoseTrack 2018, COCO 2017 (train, instance annotations), ImageNet-1k (ILSVRC 2012 classification) (Cocktail14: AI Challenger, CrowdPose, MPII, sub-JHMDB, Halpe, PoseTrack18, COCO-WholeBody, UBody, Human-Art, WFLW, 300W, COFW, LaPa and InterHand2.6M (the last six are not in the dataset list; several are research-only), distilled from RTMW-x; the CSPNeXt backbone was pre-trained on ImageNet-1k and on body / whole-body keypoints)rtmw-x: COCO-WholeBody, UBody (upper-body whole-body keypoints, from OSX), AI Challenger human keypoints (2017), CrowdPose, MPII Human Pose, sub-JHMDB (J-HMDB), Halpe-FullBody, PoseTrack 2018, COCO 2017 (train, instance annotations), ImageNet-1k (ILSVRC 2012 classification) (Cocktail14: AI Challenger, CrowdPose, MPII, sub-JHMDB, Halpe, PoseTrack18, COCO-WholeBody, UBody, Human-Art, WFLW, 300W, COFW, LaPa and InterHand2.6M (the last six are not in the dataset list; several are research-only); the CSPNeXt backbone was pre-trained on ImageNet-1k and on body / whole-body keypoints)rtmw-l-384: COCO-WholeBody, UBody (upper-body whole-body keypoints, from OSX), AI Challenger human keypoints (2017), CrowdPose, MPII Human Pose, sub-JHMDB (J-HMDB), Halpe-FullBody, PoseTrack 2018, COCO 2017 (train, instance annotations), ImageNet-1k (ILSVRC 2012 classification) (Cocktail14: AI Challenger, CrowdPose, MPII, sub-JHMDB, Halpe, PoseTrack18, COCO-WholeBody, UBody, Human-Art, WFLW, 300W, COFW, LaPa and InterHand2.6M (the last six are not in the dataset list; several are research-only), distilled from RTMW-x; the CSPNeXt backbone was pre-trained on ImageNet-1k and on body / whole-body keypoints)rtmw-x-384: COCO-WholeBody, UBody (upper-body whole-body keypoints, from OSX), AI Challenger human keypoints (2017), CrowdPose, MPII Human Pose, sub-JHMDB (J-HMDB), Halpe-FullBody, PoseTrack 2018, COCO 2017 (train, instance annotations), ImageNet-1k (ILSVRC 2012 classification) (Cocktail14: AI Challenger, CrowdPose, MPII, sub-JHMDB, Halpe, PoseTrack18, COCO-WholeBody, UBody, Human-Art, WFLW, 300W, COFW, LaPa and InterHand2.6M (the last six are not in the dataset list; several are research-only); the CSPNeXt backbone was pre-trained on ImageNet-1k and on body / whole-body keypoints)rtmpose-t-halpe26: COCO 2017 (train, instance annotations), AI Challenger human keypoints (2017), CrowdPose, MPII Human Pose, sub-JHMDB (J-HMDB), Halpe-FullBody, PoseTrack 2018, ImageNet-1k (ILSVRC 2012 classification) (body7 with the 26 Halpe keypoints (COCO-WholeBody feet annotations for COCO): COCO, AI Challenger, CrowdPose, MPII, sub-JHMDB, Halpe and PoseTrack18; the CSPNeXt backbone was pre-trained on ImageNet-1k and on body7)rtmpose-s-halpe26: COCO 2017 (train, instance annotations), AI Challenger human keypoints (2017), CrowdPose, MPII Human Pose, sub-JHMDB (J-HMDB), Halpe-FullBody, PoseTrack 2018, ImageNet-1k (ILSVRC 2012 classification) (body7 with the 26 Halpe keypoints (COCO-WholeBody feet annotations for COCO): COCO, AI Challenger, CrowdPose, MPII, sub-JHMDB, Halpe and PoseTrack18; the CSPNeXt backbone was pre-trained on ImageNet-1k and on body7)rtmpose-m-halpe26: COCO 2017 (train, instance annotations), AI Challenger human keypoints (2017), CrowdPose, MPII Human Pose, sub-JHMDB (J-HMDB), Halpe-FullBody, PoseTrack 2018, ImageNet-1k (ILSVRC 2012 classification) (body7 with the 26 Halpe keypoints (COCO-WholeBody feet annotations for COCO): COCO, AI Challenger, CrowdPose, MPII, sub-JHMDB, Halpe and PoseTrack18; the CSPNeXt backbone was pre-trained on ImageNet-1k and on body7)rtmpose-l-halpe26: COCO 2017 (train, instance annotations), AI Challenger human keypoints (2017), CrowdPose, MPII Human Pose, sub-JHMDB (J-HMDB), Halpe-FullBody, PoseTrack 2018, ImageNet-1k (ILSVRC 2012 classification) (body7 with the 26 Halpe keypoints (COCO-WholeBody feet annotations for COCO): COCO, AI Challenger, CrowdPose, MPII, sub-JHMDB, Halpe and PoseTrack18; the CSPNeXt backbone was pre-trained on ImageNet-1k and on body7)rtmpose-m-halpe26-384: COCO 2017 (train, instance annotations), AI Challenger human keypoints (2017), CrowdPose, MPII Human Pose, sub-JHMDB (J-HMDB), Halpe-FullBody, PoseTrack 2018, ImageNet-1k (ILSVRC 2012 classification) (body7 with the 26 Halpe keypoints (COCO-WholeBody feet annotations for COCO): COCO, AI Challenger, CrowdPose, MPII, sub-JHMDB, Halpe and PoseTrack18; the CSPNeXt backbone was pre-trained on ImageNet-1k and on body7)rtmpose-l-halpe26-384: COCO 2017 (train, instance annotations), AI Challenger human keypoints (2017), CrowdPose, MPII Human Pose, sub-JHMDB (J-HMDB), Halpe-FullBody, PoseTrack 2018, ImageNet-1k (ILSVRC 2012 classification) (body7 with the 26 Halpe keypoints (COCO-WholeBody feet annotations for COCO): COCO, AI Challenger, CrowdPose, MPII, sub-JHMDB, Halpe and PoseTrack18; the CSPNeXt backbone was pre-trained on ImageNet-1k and on body7)rtmpose-x-halpe26-384: COCO 2017 (train, instance annotations), AI Challenger human keypoints (2017), CrowdPose, MPII Human Pose, sub-JHMDB (J-HMDB), Halpe-FullBody, PoseTrack 2018, ImageNet-1k (ILSVRC 2012 classification) (body7 with the 26 Halpe keypoints (COCO-WholeBody feet annotations for COCO): COCO, AI Challenger, CrowdPose, MPII, sub-JHMDB, Halpe and PoseTrack18; the CSPNeXt backbone was pre-trained on ImageNet-1k and on body7)rtmpose-m-hand5: COCO-WholeBody, Halpe-FullBody, COCO 2017 (train, instance annotations), AI Challenger human keypoints (2017), ImageNet-1k (ILSVRC 2012 classification) (hand5: COCO-WholeBody-Hand, OneHand10K, FreiHAND, RHD and Halpe hands (OneHand10K, FreiHAND and RHD are not in the dataset list; research-only terms); the CSPNeXt backbone was pre-trained on ImageNet-1k and on AI Challenger + COCO body keypoints)rtmpose-t-face6: COCO-WholeBody, Halpe-FullBody, ImageNet-1k (ILSVRC 2012 classification) (face6: COCO-WholeBody-Face, WFLW, 300W, COFW, Halpe faces and LaPa, converted to the 106 LaPa landmarks (WFLW, 300W, COFW and LaPa are not in the dataset list; research-only terms); the CSPNeXt backbone was pre-trained on ImageNet-1k)rtmpose-s-face6: COCO-WholeBody, Halpe-FullBody, ImageNet-1k (ILSVRC 2012 classification) (face6: COCO-WholeBody-Face, WFLW, 300W, COFW, Halpe faces and LaPa, converted to the 106 LaPa landmarks (WFLW, 300W, COFW and LaPa are not in the dataset list; research-only terms); the CSPNeXt backbone was pre-trained on ImageNet-1k)rtmpose-m-face6: COCO-WholeBody, Halpe-FullBody, ImageNet-1k (ILSVRC 2012 classification) (face6: COCO-WholeBody-Face, WFLW, 300W, COFW, Halpe faces and LaPa, converted to the 106 LaPa landmarks (WFLW, 300W, COFW and LaPa are not in the dataset list; research-only terms); the CSPNeXt backbone was pre-trained on ImageNet-1k)rtmpose-m-ap10k: AP-10K (animal pose), COCO 2017 (train, instance annotations), AI Challenger human keypoints (2017), ImageNet-1k (ILSVRC 2012 classification) (AP-10K (animal keypoints, 54 species); the CSPNeXt backbone was pre-trained on ImageNet-1k and on AI Challenger + COCO body keypoints)
- Options:
flip_test= False - Kernel libraries:
cudnn_hacc,fvm_conv,fvm_extra,fvm_mb,fvm_r3 - Variants:
rtmpose-t,rtmpose-s,rtmpose-m,rtmpose-l,rtmpose-m-384,rtmpose-l-384,dwpose-t,dwpose-s,dwpose-m,dwpose-l,dwpose-l-384,rtmw-m,rtmw-l,rtmw-x,rtmw-l-384,rtmw-x-384,rtmpose-t-halpe26,rtmpose-s-halpe26,rtmpose-m-halpe26,rtmpose-l-halpe26,rtmpose-m-halpe26-384,rtmpose-l-halpe26-384,rtmpose-x-halpe26-384,rtmpose-m-hand5,rtmpose-t-face6,rtmpose-s-face6,rtmpose-m-face6,rtmpose-m-ap10k - Outputs: keypoints, keypoint_scores, boxes, scores; label_set:
coco-keypoints-17 - Input size: any H x W >= 32, padded to a multiple of 32
- Tested up to (portable, fp16, RTX 5060 Ti 16 GB):
rtmpose-t3000x4000 (0.4 GiB peak allocated, 0.5 reserved);rtmpose-s3000x4000 (0.4 GiB peak allocated, 0.5 reserved);rtmpose-m3000x4000 (0.5 GiB peak allocated, 0.5 reserved);rtmpose-l3000x4000 (0.5 GiB peak allocated, 0.5 reserved);rtmpose-m-3843000x4000 (0.5 GiB peak allocated, 0.6 reserved);rtmpose-l-3843000x4000 (0.6 GiB peak allocated, 1.0 reserved);dwpose-t3000x4000 (0.4 GiB peak allocated, 0.5 reserved);dwpose-s3000x4000 (0.5 GiB peak allocated, 0.5 reserved);dwpose-m3000x4000 (0.5 GiB peak allocated, 0.5 reserved);dwpose-l3000x4000 (0.5 GiB peak allocated, 0.5 reserved);dwpose-l-3843000x4000 (0.6 GiB peak allocated, 1.0 reserved);rtmw-m3000x4000 (0.5 GiB peak allocated, 0.5 reserved);rtmw-l3000x4000 (0.5 GiB peak allocated, 0.5 reserved);rtmw-x3000x4000 (0.6 GiB peak allocated, 0.7 reserved);rtmw-l-3843000x4000 (0.7 GiB peak allocated, 1.0 reserved);rtmw-x-3843000x4000 (0.8 GiB peak allocated, 1.0 reserved);rtmpose-t-halpe263000x4000 (0.4 GiB peak allocated, 0.5 reserved);rtmpose-s-halpe263000x4000 (0.4 GiB peak allocated, 0.5 reserved);rtmpose-m-halpe263000x4000 (0.5 GiB peak allocated, 0.5 reserved);rtmpose-l-halpe263000x4000 (0.5 GiB peak allocated, 0.5 reserved);rtmpose-m-halpe26-3843000x4000 (0.5 GiB peak allocated, 0.6 reserved);rtmpose-l-halpe26-3843000x4000 (0.6 GiB peak allocated, 1.0 reserved);rtmpose-x-halpe26-3843000x4000 (0.8 GiB peak allocated, 0.9 reserved);rtmpose-m-hand53000x4000 (0.5 GiB peak allocated, 0.5 reserved);rtmpose-t-face63000x4000 (0.4 GiB peak allocated, 0.5 reserved);rtmpose-s-face63000x4000 (0.5 GiB peak allocated, 0.5 reserved);rtmpose-m-face63000x4000 (0.5 GiB peak allocated, 0.5 reserved);rtmpose-m-ap10k3000x4000 (0.5 GiB peak allocated, 0.5 reserved) rtmpose-t1080x1920, first call: portable + torch.compile 11 s (cold Inductor cache), TensorRT fp16 16 s (ONNX + engine).rtmpose-s1080x1920, first call: portable + torch.compile 11 s (cold Inductor cache), TensorRT fp16 16 s (ONNX + engine).rtmpose-m1080x1920, first call: portable + torch.compile 13 s (cold Inductor cache), TensorRT fp16 18 s (ONNX + engine).rtmpose-l1080x1920, first call: portable + torch.compile 14 s (cold Inductor cache), TensorRT fp16 19 s (ONNX + engine).rtmpose-m-3841080x1920, first call: portable + torch.compile 14 s (cold Inductor cache), TensorRT fp16 19 s (ONNX + engine).rtmpose-l-3841080x1920, first call: portable + torch.compile 15 s (cold Inductor cache), TensorRT fp16 20 s (ONNX + engine).dwpose-t1080x1920, first call: portable + torch.compile 11 s (cold Inductor cache), TensorRT fp16 16 s (ONNX + engine).dwpose-s1080x1920, first call: portable + torch.compile 12 s (cold Inductor cache), TensorRT fp16 17 s (ONNX + engine).dwpose-m1080x1920, first call: portable + torch.compile 13 s (cold Inductor cache), TensorRT fp16 19 s (ONNX + engine).dwpose-l1080x1920, first call: portable + torch.compile 15 s (cold Inductor cache), TensorRT fp16 20 s (ONNX + engine).dwpose-l-3841080x1920, first call: portable + torch.compile 15 s (cold Inductor cache), TensorRT fp16 21 s (ONNX + engine).rtmw-m1080x1920, first call: portable + torch.compile 19 s (cold Inductor cache), TensorRT fp16 23 s (ONNX + engine).rtmw-l1080x1920, first call: portable + torch.compile 19 s (cold Inductor cache), TensorRT fp16 25 s (ONNX + engine).rtmw-x1080x1920, first call: portable + torch.compile 20 s (cold Inductor cache), TensorRT fp16 27 s (ONNX + engine).rtmw-l-3841080x1920, first call: portable + torch.compile 20 s (cold Inductor cache), TensorRT fp16 27 s (ONNX + engine).rtmw-x-3841080x1920, first call: portable + torch.compile 21 s (cold Inductor cache), TensorRT fp16 30 s (ONNX + engine).rtmpose-t-halpe261080x1920, first call: portable + torch.compile 13 s (cold Inductor cache), TensorRT fp16 17 s (ONNX + engine).rtmpose-s-halpe261080x1920, first call: portable + torch.compile 12 s (cold Inductor cache), TensorRT fp16 17 s (ONNX + engine).rtmpose-m-halpe261080x1920, first call: portable + torch.compile 13 s (cold Inductor cache), TensorRT fp16 18 s (ONNX + engine).rtmpose-l-halpe261080x1920, first call: portable + torch.compile 14 s (cold Inductor cache), TensorRT fp16 20 s (ONNX + engine).rtmpose-m-halpe26-3841080x1920, first call: portable + torch.compile 14 s (cold Inductor cache), TensorRT fp16 19 s (ONNX + engine).rtmpose-l-halpe26-3841080x1920, first call: portable + torch.compile 15 s (cold Inductor cache), TensorRT fp16 21 s (ONNX + engine).rtmpose-x-halpe26-3841080x1920, first call: portable + torch.compile 17 s (cold Inductor cache), TensorRT fp16 23 s (ONNX + engine).rtmpose-m-hand51080x1920, first call: portable + torch.compile 13 s (cold Inductor cache), TensorRT fp16 18 s (ONNX + engine).rtmpose-t-face61080x1920, first call: portable + torch.compile 11 s (cold Inductor cache), TensorRT fp16 16 s (ONNX + engine).rtmpose-s-face61080x1920, first call: portable + torch.compile 12 s (cold Inductor cache), TensorRT fp16 17 s (ONNX + engine).rtmpose-m-face61080x1920, first call: portable + torch.compile 13 s (cold Inductor cache), TensorRT fp16 18 s (ONNX + engine).rtmpose-m-ap10k1080x1920, first call: portable + torch.compile 13 s (cold Inductor cache), TensorRT fp16 18 s (ONNX + engine).rtmpose-toriginal (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-toriginal 1080x1920: torch.autocast(fp16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-soriginal (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-soriginal 1080x1920: torch.autocast(fp16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-moriginal (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-moriginal 1080x1920: torch.autocast(fp16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-loriginal (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-loriginal 1080x1920: torch.autocast(fp16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-m-384original (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-m-384original 1080x1920: torch.autocast(fp16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-l-384original (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-l-384original 1080x1920: torch.autocast(fp16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed calldwpose-toriginal (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed calldwpose-toriginal 1080x1920: torch.autocast(fp16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed calldwpose-soriginal (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed calldwpose-soriginal 1080x1920: torch.autocast(fp16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed calldwpose-moriginal (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed calldwpose-moriginal 1080x1920: torch.autocast(fp16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed calldwpose-loriginal (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed calldwpose-loriginal 1080x1920: torch.autocast(fp16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed calldwpose-l-384original (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed calldwpose-l-384original 1080x1920: torch.autocast(fp16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmw-moriginal (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmw-moriginal 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmw-loriginal (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmw-loriginal 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmw-xoriginal (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmw-xoriginal 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmw-l-384original (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmw-l-384original 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmw-x-384original (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmw-x-384original 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-t-halpe26original (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-t-halpe26original 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-s-halpe26original (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-s-halpe26original 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-m-halpe26original (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-m-halpe26original 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-l-halpe26original (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-l-halpe26original 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-m-halpe26-384original (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-m-halpe26-384original 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-l-halpe26-384original (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-l-halpe26-384original 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-x-halpe26-384original (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-x-halpe26-384original 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-m-hand5original (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-m-hand5original 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-t-face6original (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-t-face6original 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-s-face6original (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-s-face6original 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-m-face6original (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-m-face6original 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-m-ap10koriginal (fp32) 1080x1920: mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-m-ap10koriginal 1080x1920: torch.autocast(float16) around the call; mmpose 1.3.2 inference_topdown (CPU cv2 crop, network, SimCC decode; flip_test off), all in the timed callrtmpose-tload / first call: original 0.13 s / 0.152 s, portable 0.1 s / 0.879 s, kernels 1.42 s / 0.291 srtmpose-sload / first call: original 0.16 s / 0.164 s, portable 0.1 s / 1.033 s, kernels 1.4 s / 0.291 srtmpose-mload / first call: original 0.23 s / 0.156 s, portable 0.22 s / 1.105 s, kernels 1.5 s / 0.291 srtmpose-lload / first call: original 0.38 s / 0.179 s, portable 0.94 s / 0.858 s, kernels 1.6 s / 0.29 srtmpose-m-384load / first call: original 0.23 s / 0.349 s, portable 0.2 s / 1.245 s, kernels 1.48 s / 0.29 srtmpose-l-384load / first call: original 0.58 s / 0.165 s, portable 0.41 s / 1.0 s, kernels 1.56 s / 0.291 sdwpose-tload / first call: original 0.16 s / 0.326 s, portable 0.17 s / 1.243 s, kernels 0.69 s / 0.29 sdwpose-sload / first call: original 0.19 s / 0.47 s, portable 0.12 s / 1.022 s, kernels 1.48 s / 0.29 sdwpose-mload / first call: original 0.26 s / 0.596 s, portable 0.24 s / 1.099 s, kernels 1.51 s / 0.292 sdwpose-lload / first call: original 0.5 s / 1.143 s, portable 0.88 s / 1.831 s, kernels 1.64 s / 0.28 sdwpose-l-384load / first call: original 0.42 s / 0.425 s, portable 0.58 s / 1.102 s, kernels 1.67 s / 0.285 srtmw-mload / first call: original 0.54 s / 1.213 s, portable 0.32 s / 0.896 s, kernels 1.02 s / 0.301 srtmw-lload / first call: original 0.76 s / 0.632 s, portable 0.39 s / 1.032 s, kernels 1.72 s / 0.306 srtmw-xload / first call: original 1.23 s / 0.64 s, portable 1.35 s / 0.833 s, kernels 2.1 s / 0.283 srtmw-l-384load / first call: original 0.75 s / 0.56 s, portable 0.34 s / 1.198 s, kernels 1.69 s / 0.279 srtmw-x-384load / first call: original 1.05 s / 0.419 s, portable 0.92 s / 0.97 s, kernels 1.36 s / 0.329 srtmpose-t-halpe26load / first call: original 0.18 s / 0.707 s, portable 0.16 s / 0.872 s, kernels 0.73 s / 0.292 srtmpose-s-halpe26load / first call: original 0.19 s / 0.614 s, portable 0.07 s / 0.943 s, kernels 1.46 s / 0.296 srtmpose-m-halpe26load / first call: original 0.25 s / 0.445 s, portable 0.14 s / 1.037 s, kernels 1.51 s / 0.297 srtmpose-l-halpe26load / first call: original 0.37 s / 0.451 s, portable 0.25 s / 1.128 s, kernels 1.55 s / 0.292 srtmpose-m-halpe26-384load / first call: original 0.26 s / 0.341 s, portable 0.13 s / 1.15 s, kernels 1.48 s / 0.292 srtmpose-l-halpe26-384load / first call: original 0.41 s / 0.382 s, portable 0.26 s / 1.239 s, kernels 1.57 s / 0.29 srtmpose-x-halpe26-384load / first call: original 0.89 s / 0.938 s, portable 0.41 s / 1.17 s, kernels 1.76 s / 0.276 srtmpose-m-hand5load / first call: original 0.23 s / 0.219 s, portable 0.13 s / 1.134 s, kernels 1.49 s / 0.289 srtmpose-t-face6load / first call: original 0.18 s / 0.339 s, portable 0.16 s / 1.295 s, kernels 0.71 s / 0.292 srtmpose-s-face6load / first call: original 0.31 s / 0.354 s, portable 0.17 s / 1.11 s, kernels 0.69 s / 0.296 srtmpose-m-face6load / first call: original 0.29 s / 0.203 s, portable 0.15 s / 1.146 s, kernels 1.49 s / 0.289 srtmpose-m-ap10kload / first call: original 0.23 s / 0.19 s, portable 0.13 s / 1.139 s, kernels 1.48 s / 0.291 s- Protocol, first call, 3 warm-up, peak reset, median of 3 rounds x median of 5
Measured
Top-down on the same every-tenth-image subset for all modes, ground-truth person boxes, no flip test; people ranked by the mean of their first 17 keypoint scores; pycocotools OKS AP on COCO body joints. Top-down on the ground-truth person boxes (non-crowd, at least one labelled keypoint), no flip test; people ranked by the mean of their 17 keypoint scores (not mmpose's rescoring and OKS-NMS), so not the papers' AP; pycocotools OKS AP; whole-body models scored on their 17 body keypoints. One prediction per annotated hand box. PCK is the mean across landmarks of the fraction within 20% of the longer box side. EPE is the mean pixel distance across annotated landmarks. Flip testing is disabled. Original = the original code named under Details, same checkpoint and protocol.
| variant | dataset | original fp32 | original fp16 | portable | portable fp32 | kernels | note |
|---|---|---|---|---|---|---|---|
rtmpose-t | COCO val2017, ground-truth person boxes, keypoint AP | 65.2 | 65.1 | 65.2 | 65.2 | ||
rtmpose-s | COCO val2017, ground-truth person boxes, keypoint AP | 69.8 | 69.8 | 69.8 | 69.8 | ||
rtmpose-m | COCO val2017, ground-truth person boxes, keypoint AP | 75.6 | 75.6 | 75.6 | 75.6 | ||
rtmpose-l | COCO val2017, ground-truth person boxes, keypoint AP | 77.9 | 77.9 | 77.9 | 78.0 | ||
rtmpose-m-384 | COCO val2017, ground-truth person boxes, keypoint AP | 78.0 | 77.9 | 77.9 | 78.0 | ||
rtmpose-l-384 | COCO val2017, ground-truth person boxes, keypoint AP | 80.0 | 80.0 | 80.0 | 79.9 | ||
dwpose-t | COCO val2017, ground-truth person boxes, keypoint AP | 60.2 | 60.2 | 60.1 | 60.1 | ||
dwpose-s | COCO val2017, ground-truth person boxes, keypoint AP | 65.6 | 65.6 | 65.6 | 65.5 | ||
dwpose-m | COCO val2017, ground-truth person boxes, keypoint AP | 71.7 | 71.8 | 71.8 | 71.8 | ||
dwpose-l | COCO val2017, ground-truth person boxes, keypoint AP | 73.9 | 73.9 | 74.0 | 74.0 | ||
dwpose-l-384 | COCO val2017, ground-truth person boxes, keypoint AP | 75.6 | 75.6 | 75.7 | 75.6 | ||
rtmw-m | COCO val2017, ground-truth person boxes, keypoint AP | 74.5 | 74.5 | 74.5 | 74.6 | ||
rtmw-l | COCO val2017, ground-truth person boxes, keypoint AP | 76.9 | 76.9 | 77.0 | 76.9 | ||
rtmw-x | COCO val2017, ground-truth person boxes, keypoint AP | 77.3 | 77.4 | 77.3 | 77.4 | ||
rtmw-l-384 | COCO val2017, every 10th image, GT person boxes, keypoint AP | 80.2 | 80.3 | 80.2 | 80.1 | 80.1 | |
rtmw-l-384 | COCO val2017, ground-truth person boxes, keypoint AP | 79.4 | 79.4 | 79.5 | 79.5 | ||
rtmw-x-384 | COCO val2017, every 10th image, GT person boxes, keypoint AP | 80.3 | 80.2 | 80.1 | 80.3 | 80.4 | |
rtmw-x-384 | COCO val2017, ground-truth person boxes, keypoint AP | 79.6 | 79.6 | 79.6 | 79.6 | ||
rtmpose-t-halpe26 | COCO val2017, ground-truth person boxes, keypoint AP | 65.3 | 65.5 | 65.3 | 65.3 | ||
rtmpose-s-halpe26 | COCO val2017, every 10th image, GT person boxes, keypoint AP | 71.4 | 71.3 | 71.4 | 71.2 | ||
rtmpose-s-halpe26 | COCO val2017, ground-truth person boxes, keypoint AP | 70.0 | 70.0 | 69.9 | 70.0 | ||
rtmpose-m-halpe26 | COCO val2017, ground-truth person boxes, keypoint AP | 76.3 | 76.2 | 76.2 | 76.2 | ||
rtmpose-l-halpe26 | COCO val2017, ground-truth person boxes, keypoint AP | 78.5 | 78.5 | 78.5 | 78.5 | ||
rtmpose-m-halpe26-384 | COCO val2017, ground-truth person boxes, keypoint AP | 78.3 | 78.4 | 78.3 | 78.3 | ||
rtmpose-l-halpe26-384 | COCO val2017, ground-truth person boxes, keypoint AP | 79.8 | 79.8 | 79.8 | 80.0 | ||
rtmpose-x-halpe26-384 | COCO val2017, ground-truth person boxes, keypoint AP | 80.5 | 80.6 | 80.5 | 80.4 | ||
rtmpose-m-hand5 | COCO-WholeBody validation (1,568 images), hand [email protected] % / EPE in px | 80.98 / 4.56 |
Measured on portable: NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks 2241b095+uncommitted / be54b8b0, 2026-11-09 to 2026-11-11; portable32: NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks 3bf0bac7 / 44640884, 2026-11-18; kernels: NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks 2241b095+uncommitted / be54b8b0, 2026-11-09 to 2026-11-11.
Parity
CPU, float32: max |diff| <= 0.002 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.66 and p99 <= 3.6 (absolute error of the output keypoints).
| variant | kernels error vs original, mean / p99 | GPU, torch | kernels used |
|---|---|---|---|
rtmpose-t | 0.40835 / 2.02585 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-s | 0.35962 / 2.04983 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-m | 0.14799 / 1.19873 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-l | 0.19814 / 1.19873 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-m-384 | 0.32639 / 2.09004 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-l-384 | 0.2774 / 0.79916 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
dwpose-t | 0.18757 / 1.19873 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
dwpose-s | 0.21925 / 1.19873 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
dwpose-m | 0.09716 / 1.19873 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
dwpose-l | 0.16913 / 1.64226 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
dwpose-l-384 | 0.16432 / 2.22965 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmw-m | 0.10568 / 1.19873 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmw-l | 0.07894 / 1.19873 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmw-x | 0.12505 / 1.19873 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmw-l-384 | 0.94198 / 1.29593 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmw-x-384 | 0.68362 / 2.91205 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-t-halpe26 | 0.51209 / 4.77095 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-s-halpe26 | 1.41323 / 27.36431 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-m-halpe26 | 0.24452 / 1.78611 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-l-halpe26 | 0.20565 / 1.19873 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-m-halpe26-384 | 0.16174 / 1.48645 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-l-halpe26-384 | 0.14576 / 0.79916 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-x-halpe26-384 | 0.15131 / 0.79918 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-m-hand5 | 0.39958 / 2.61323 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-t-face6 | 0.12106 / 1.19873 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-s-face6 | 0.10929 / 1.19873 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-m-face6 | 0.07662 / 1.19873 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
rtmpose-m-ap10k | 0.52728 / 4.73499 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fvm_conv@f1d064e539e1:sm_120, fvm_extra@3208ce525a91:sm_120, fvm_mb@720dd3f29bf7:sm_120, fvm_r3@f54eb890e23c:sm_120 |
Example output
One fixed input through this checkpoint, the same for every model of the task (kernels · float16 · NVIDIA GeForce RTX 5060 Ti).
Input: Astronaut Eileen Collins, NASA (public domain).