Retrieval models
2 retrieval models. Each section says what the model is, whether it may be used commercially and how fast it is; licences and checks are under Details.
Input: one RGB image. Output: a global descriptor [D] (unit L2 norm) for image retrieval / visual place recognition; fastforks.retrieval.top_k and pairs find the nearest images (cosine similarity) on the GPU.
Measured on RTX 5060 Ti, torch 2.14.1+cu130; fastforks bench MODEL --size HxW measures yours. The summary uses one input, 480x640: kernels time with its speed-up over the faster original run, and kernels peak reserved memory with its ratio to the smaller original one. Where measured (one variant per family), a variant's table also lists portable under torch.compile and the fp16 TensorRT engine of portable (export; memory: the engine's device memory + torch's peak). – = not measured; no kernels = the family has no kernels backend (reason under its table). Reserved memory is the allocator's cache: kernels (CUDA graphs, cuDNN workspaces) can reserve a few percent to some tens of percent more than portable at the same allocated peak.
Other devices: Apple M1 Pro 16 GB, NVIDIA L4, NVIDIA L40S, NVIDIA A10, NVIDIA A100-SXM4-40GB, NVIDIA RTX PRO 6000; across devices.
| model | commercial | kernels time (ms), speed-up vs original | kernels peak memory (MiB, reserved), ratio to original |
|---|---|---|---|
boq-dinov2 | yes | 2.84, 2.6x | 256, 1.0x |
megaloc | yes | 4.26, 2.3x | 508, 0.37x |
All models
Modes, licences, training data and outputs per model
Commercial status per variant: yes if the code and the weights licence both allow commercial use (the training data is listed but does not change it; weights without a licence of their own are assumed to carry the code licence); no if either forbids it; unclear for custom licence terms (treat as no).
| model | modes | code licence | weights licence | training data | commercial | outputs |
|---|---|---|---|---|---|---|
boq-dinov2 | portable, kernels | MIT | MIT | lvd-142m, gsv-cities | yes | descriptor |
megaloc | portable, kernels | MIT | MIT | lvd-142m, gsv-cities, msls, megascenes, scannet, sf-xl | yes | descriptor |
Training datasets
| id | dataset | licence | commercial use |
|---|---|---|---|
lvd-142m | LVD-142M (DINOv2's curated web-image set, not released) | not stated | unclear |
gsv-cities | GSV-Cities (Google Street View images of 62k places in 40 cities) | not stated | unclear |
msls | Mapillary Street-Level Sequences | CC-BY-NC-SA-4.0 | no |
megascenes | MegaScenes (Internet photo collections of scenes from Wikimedia Commons) | not stated | unclear |
scannet | ScanNet (RGB-D scans of indoor scenes) | LicenseRef-noncommercial | no |
sf-xl | San Francisco eXtra Large (street-view images of San Francisco) | not stated | unclear |