ReSplat
ReSplat (ETH Zurich / cvg): 3D Gaussian splats from several posed images, refined by rendering them back; saves a 3DGS .ply. MIT (DL3DV Small), non-commercial (Base).
experimental Fork of cvg/resplat@cae7ddc · backends: portable, kernels
- Paper ReSplat: Learning Recurrent Gaussian Splats arXiv · first released 9 Oct 2025
Paper 2025 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 multiview
model = multiview.load("resplat-small-dl3dv") # kernels on a supported GPU, else portable PyTorch
result = model(images)
result.depth # also: confidence, gaussians
result.backend # what ran: mode, device, dtype, kernels fastforks bench resplat-small-dl3dv 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, 480x640
| Backend | dtype | Time per call (ms) | vs original fp16 | Peak memory allocated / reserved | Load (s) | First call (s) |
|---|---|---|---|---|---|---|
| Original fp32 | fp32 | not measured | – | – | – | – |
| Original (own default)the authors' mixed-precision option (bfloat16); cvg/resplat@cae7ddc4cdbd EncoderReSplat, pointops + gsplat extensions, 2 refinement steps, released bfloat16 autocast | bf16 | 493 | – | 5,148 MiB/5,376 MiB | 1.0 | 2.68 |
| Portable fp32 | fp32 | not measured | – | – | – | – |
| Portable | fp16 | 1723 | 0.29× | 4,995 MiB/5,180 MiB | 0.7 | 2.51 |
| Kernels | fp16 | 362 | 1.4× | 4,981 MiB/5,148 MiB | 3.9 | 2.84 |
| Portable + torch.compile | 1538 | 0.32× | 4,995 MiB/5,138 MiB | 243 (build) | – | |
| TensorRTplanned for 0.2.1 | fails | – | – | – | – |
Measured on RTX 5060 Ti 16 GB, torch 2.14.1+cu130, 2026-11-15 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. 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, 2026-11-17 | Portable | fp32 | 14422 | 11.8 GiB process | skip | 480x640 |
| Kernels | fp16 | not implemented | – | 480x640 | ||
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
- MIT LICENSE
- Weights
- MIT source
- Training data
-
- DL3DV-10K (multi-view real-world scenes for 3D vision) CC-BY-NC-4.0, non-commercial
- Code and checkpoints are MIT. The Base RealEstate10K checkpoint's monocular branch is initialised from Depth Anything V2 Base (CC-BY-NC-4.0) and is labelled non-commercial; the Small DL3DV checkpoint builds on Depth Anything V2 Small (Apache-2.0).
Our reading of the licences, not legal advice.
Weights
resplat-small-dl3dv-256x448-view8-548993fe.pth 307 MB· hosted on Hugging Facefastforks downloads and checks these files on first use (fastforks download resplat-small-dl3dv).
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.
DTU scans 1
9, 24, 33, 75, view 24 + 7 sources, AbsRel / delta1 %
| Run | dtype | AbsRel ↓ | δ1.25 % ↑ | MAE mm ↓ | scans | ms / scan |
|---|---|---|---|---|---|---|
| Original fp32 | not measured | |||||
| Original fp16 | not measured | |||||
| Portable fp32 | fp32 | 0.4233 | 10.62 | 270.85 | 5 | 1466 |
| Portable | not measured | |||||
| Kernels | not measured | |||||
The original code was not run on this dataset, so the table cannot show the distance to it.
640x512 crops, the 7 most co-visible views as sources; GT = depth rendered from the structured-light points. AbsRel / delta1 over GT pixels, after per-image median scaling unless the model takes the cameras (calibrated MVS: cameras, error in mm; PatchMatchNet also the depth range 425-935 mm, MVSAnywhere its own range from the camera baselines). Measured on RTX 5060 Ti, torch 2.14.1+cu130, fastforks 30c7a8e9, 2026-11-18.
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| <= 5e-05 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.01 and p99 <= 0.05 (absolute error of the output depth).
| Check | Tolerance | Measured |
|---|---|---|
| CPU fp32, max |diff| | ≤ 0.00005 | not recorded |
| cpu_outliers | ≤ 0.02 | not recorded |
| GPU default dtypes, mean |diff| | ≤ 0.01 | 0.00302 |
| GPU default dtypes, p99 |diff| | ≤ 0.05 | 0.00738 |
Recorded for resplat-small-dl3dv on RTX 5060 Ti, torch 2.14.1+cu130, 2026-11-14, with these kernels: cudnn_haccfastbbgsraster
Details
ReSplat (Xu et al. 2025): feed-forward 3D Gaussian splatting from posed, calibrated images with recurrent refinement. DepthSplat's multi-view depth network (cost volume + DINOv2) and a kNN point transformer predict latent Gaussians (SH degree 3); refinement steps render the input views, compare them with the images (ResNet-18 features + colour) and update every Gaussian with a second point transformer. Returns every view's depth and the Gaussians in the given world frame (`result.gaussians`, `.save("scene.ply")`).
- Input / output
- Input: several images of one scene (posed and calibrated for multi-view stereo, unposed otherwise). Output: depth maps and, where the model predicts them, cameras and 3-D points.
- Input size
- any H x W >= 1, resized or padded inside the model; tested up to 3000x4000 (6,071 MiB allocated)
- Options
long_side=0>= 0· changes the outputrefine=2>= 0· changes the output
- Outputs
depth,confidence,gaussians- Checkpoint
vit="vits",channels=256,latent=4,fixed_latent=true,gaussians_per_point=1,same_points=false,lowest=8,train_size=[256,448],far=200- CUDA kernels
fastbbfast backbone kernels: flash attention, window attention, instance norm and rotary embeddings for ViT and similar backbonescudnn_haccconvolutions with fp16 accumulation (cuDNN)gsraster3D Gaussian splatting rasteriser
- Apple (MPS)
- fp32
Limitations
- No accuracy measurement: a metric model at RealEstate10K / DL3DV scale; the only multi-view eval set here (DTU, millimetre scale) is out of its range, and the original could not run here (its gsplat rasteriser module is not installable), so no tier has a published-protocol score.
- Portable runs upstream's two CUDA extensions (kNN, gsplat rasteriser) as PyTorch ops: 2.8x (base) / 3.5x (small) slower than the bf16 original at 480x640. Slow: kNN topk ~28 %, compositing cumprod/pow ~30 % of GPU time. Kernels: 2.0x / 1.4x faster.
- The portable kNN pooling calls torch ops that return non-Tensors: 38 graph breaks (`torch.compile` runs with breaks, no `torch.export`).
Full reference, generated from the repository
ReSplat (ETH Zurich / cvg): 3D Gaussian splats from several posed images, refined by rendering them back; saves a 3DGS .ply. MIT (DL3DV Small), non-commercial (Base).
Commercial use: resplat-small-dl3dv: yes; resplat-base-re10k: no (weights: CC-BY-NC-4.0).
| variant | input | backend | time (ms) | peak memory (MiB, allocated / reserved) |
|---|---|---|---|---|
resplat-small-dl3dv | synthetic (480x640) | original portable kernels portable + torch.compile TensorRT fp16 | 493 1723 362 1538 planned for 0.2.1 | 5148 / 5376 4995 / 5180 4981 / 5148 4995 / 5138 – |
resplat-base-re10k | synthetic (480x640) | original portable kernels portable + torch.compile TensorRT fp16 | 2521 7086 1280 fails: timeout after 900 s (audit 2026-11-16, RTX 5060 Ti: compile fails: TypeError: _Base.forward() missing 1 required positional arg | 10139 / 12684 6951 / 8736 6944 / 8706 – – |
Details
- Original code: https://github.com/cvg/resplat at
cae7ddc - Code licence: MIT (permissive); files:
LICENSE - Weights licence: MIT
resplat-base-re10k: CC-BY-NC-4.0
- Licence evidence: https://github.com/cvg/resplat/blob/cae7ddc/LICENSE and https://github.com/cvg/resplat/blob/cae7ddc/MODEL_ZOO.md and https://huggingface.co/haofeixu/resplat/tree/c69ad72a49b8358ac46599cf74468090425ed9c7
- Training data: DL3DV-10K (multi-view real-world scenes for 3D vision) (DL3DV-10K (non-commercial research licence), from DepthSplat / Depth Anything V2 Small depth initialisations.)
resplat-base-re10k: RealEstate10K (camera trajectories of YouTube real-estate videos) (RealEstate10K (YouTube videos, camera poses from ORB-SLAM; research use per its release), from a DepthSplat / Depth Anything V2 Base (CC-BY-NC-4.0) depth initialisation.)
- Licence note: Code and checkpoints are MIT. The Base RealEstate10K checkpoint's monocular branch is initialised from Depth Anything V2 Base (CC-BY-NC-4.0) and is labelled non-commercial; the Small DL3DV checkpoint builds on Depth Anything V2 Small (Apache-2.0).
- Options:
long_side= 0 (>= 0),refine= 2 (>= 0) - Kernel libraries:
fastbb,cudnn_hacc,gsraster - Variants:
resplat-small-dl3dv,resplat-base-re10k - Not in this release (planned for 0.2.1):
resplat-small-dl3dvONNX / TensorRT export: torch.export cannot capture data-dependent Gaussian renderer shapes - Outputs: depth, confidence, gaussians; kind:
metric - Input size: any H x W >= 1, resized or padded inside the model
- Tested up to (portable, fp16, RTX 5060 Ti 16 GB):
resplat-small-dl3dv3000x4000 (5.9 GiB peak allocated, 7.0 reserved);resplat-base-re10k3000x4000 (7.8 GiB peak allocated, 11.2 reserved) resplat-small-dl3dv480x640, first call: portable + torch.compile 243 s (cold Inductor cache).resplat-small-dl3dvoriginal 480x640: the authors' mixed-precision option (bfloat16); cvg/resplat@cae7ddc4cdbd EncoderReSplat, pointops + gsplat extensions, 2 refinement steps, released bfloat16 autocastresplat-base-re10koriginal 480x640: the authors' mixed-precision option (bfloat16); cvg/resplat@cae7ddc4cdbd EncoderReSplat, pointops + gsplat extensions, 2 refinement steps, released bfloat16 autocastresplat-small-dl3dvload / first call: original 16.8 s / 10.688 s (480x640), portable 0.72 s / 2.51 s (480x640), kernels 3.88 s / 2.839 s (480x640)resplat-base-re10kload / first call: original 2.26 s / 2.743 s (480x640), portable 0.9 s / 7.756 s (480x640), kernels 3.3 s / 1.98 s (480x640)- Protocol, first call, 3 warm-up, peak reset, median of 3 rounds x median of 5
Measured
640x512 crops, the 7 most co-visible views as sources; GT = depth rendered from the structured-light points. AbsRel / delta1 over GT pixels, after per-image median scaling unless the model takes the cameras (calibrated MVS: cameras, error in mm; PatchMatchNet also the depth range 425-935 mm, MVSAnywhere its own range from the camera baselines). Original = the original code named under Details, same checkpoint and protocol.
| variant | dataset | portable fp32 | note |
|---|---|---|---|
resplat-small-dl3dv | DTU scans 1, 9, 24, 33, 75, view 24 + 7 sources, AbsRel / delta1 % / MAE mm | 0.4233 / 10.62 / 270.85 | |
resplat-base-re10k | DTU scans 1, 9, 24, 33, 75, view 24 + 7 sources, AbsRel / delta1 % / MAE mm | 0.1396 / 98.14 / 92.13 |
Measured on NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks 30c7a8e9, 2026-11-18.
Parity
CPU, float32: max |diff| <= 5e-05 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.01 and p99 <= 0.05 (absolute error of the output depth).
| variant | kernels error vs original, mean / p99 | GPU, torch | kernels used |
|---|---|---|---|
resplat-small-dl3dv | 0.00302 / 0.00738 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fastbb@7d058322634d:sm_120, gsraster@be9d99ba7071:sm_120 |
resplat-base-re10k | 0.00424 / 0.01598 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, fastbb@7d058322634d:sm_120, gsraster@be9d99ba7071:sm_120 |
Limitations
- No accuracy measurement: a metric model at RealEstate10K / DL3DV scale; the only multi-view eval set here (DTU, millimetre scale) is out of its range, and the original could not run here (its gsplat rasteriser module is not installable), so no tier has a published-protocol score.
- Portable runs upstream's two CUDA extensions (kNN, gsplat rasteriser) as PyTorch ops: 2.8x (base) / 3.5x (small) slower than the bf16 original at 480x640. Slow: kNN topk ~28 %, compositing cumprod/pow ~30 % of GPU time. Kernels: 2.0x / 1.4x faster.
- The portable kNN pooling calls torch ops that return non-Tensors: 38 graph breaks (
torch.compileruns with breaks, notorch.export).
Example output
No example rendered for this checkpoint yet.