--- library_name: "fastforks" pipeline_tag: "image-to-image" license: "apache-2.0" tags: - "edges" - "edge-detection" - "line-art" - "controlnet-preprocessor" - "cuda" - "pytorch" - "fastforks" - "fastforks-card" --- # M-LSD line segments (fastforks) Mobile line segment detector (Gu et al., AAAI 2022; controlnet_aux `MLSDdetector`, ComfyUI's "M-LSD Lines"): a MobileNetV2 network that predicts segment centers and end-point displacements; returns the straight segments and the 1-pixel line map ControlNet's MLSD model takes. This repo is the model card only: it holds no weights. fastforks downloads the authors' weights from the links below and checks them against the SHA-256 in the model card. Code: . Kernels for the `kernels` backend: . Original code: at `b1aac318b15c`. Task: edges. Tier: experimental. Outputs: `edges`, `lines`. Backends: `portable`. Input: one RGB image. Output: a line / edge map `[H, W]` in [0, 1] at the input size (1 = line), the ControlNet preprocessors (line art, HED, PiDiNet, TEED / Anyline); M-LSD also returns straight line segments. ## Variants Weights are the authors' files (not hosted here). SHA-256 shortened to 12 hex digits; the full hashes are in [`card.toml`](https://github.com/kleinicke/fastforks/blob/main/src/fastforks/edges/models/mlsd/card.toml). | variant | weights (author link) | SHA-256 | weights licence | commercial use | |---|---|---|---|---| | `mlsd@1` | [mlsd_large_512_fp32.pth](https://raw.githubusercontent.com/lhwcv/mlsd_pytorch/2312205254e66911703decf775f626995d260f17/models/mlsd_large_512_fp32.pth) | `5696f168eb2c` | Apache-2.0 | yes | ## Licence Original-code strings, recorded as they are (research-use package; check them before any other use). Code: Apache-2.0. Licence files: `LICENSE`. Training data: Not stated for this checkpoint; the M-LSD paper trains on the Wireframe dataset (Huang et al. 2018, no licence stated).. Licence evidence: https://github.com/lhwcv/mlsd_pytorch/blob/2312205254e66911703decf775f626995d260f17/LICENSE ; https://github.com/lhwcv/mlsd_pytorch/tree/2312205254e66911703decf775f626995d260f17/models ## Input size any H x W >= 32, padded to a multiple of 32 Tested up to (portable, fp16, RTX 5060 Ti 16 GB): `mlsd` 1080x1920 (0.1 GiB peak allocated, 0.2 reserved) ## Speed and memory ### Edges, synthetic, 1080x1920 | model | time (ms): original | portable | peak memory (MiB, allocated / reserved): original | portable | | --- | --- | --- | --- | --- | | `mlsd` | 35.6 | 9.33 | 185 / 212 | 127 / 170 | ### Startup and first call | model | load (s): original | portable | kernels | first call (s): original | portable | kernels | | --- | --- | --- | --- | --- | --- | --- | | `mlsd` | 0.21 | 0.17 | – | 0.06 | 0.01 | – | ### torch.compile and TensorRT | model | input | time: portable | compile | TensorRT fp16 | kernels | preparation: compile (cold first call) | ONNX + engine build | peak MiB (alloc / reserved): portable | compile | kernels | TensorRT (engine + torch) | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | `mlsd` | 1024x1024 | 7.65 ms | 4.61 ms | planned for 0.2.1 | – | 11.4 s | – | 72 / 90 | 101 / 130 | – | – | Fixed-size cells: time (ms) / peak allocated / peak reserved memory (MiB). Hardware, protocol and caveats: [benchmarks](https://github.com/kleinicke/fastforks/blob/main/docs/benchmarks.md) (RTX 5060 Ti 16 GB, sm_120). Original = the original code named above at the precision of portable (fp16, or `(fp32)` where fp16 is unsupported or not finite); original fp32 = its default fp32 run. Peak memory is the torch allocator's. ## Accuracy ### Parity with the original code Every variant is checked against reference outputs of the original code at its pinned commit. CPU, float32: max |diff| <= 0.002 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.02 and p99 <= 0.25 (absolute error of the output `edges`). ## Usage ```python from fastforks import edges model = edges.load("mlsd") # mode="auto": the preferred backend from recorded measurements result = model("photo.jpg") result.edges # [H, W] float32 in [0, 1], 1 = line, input resolution result.backend # what ran, e.g. Backend(mode='portable', device='cuda:0', dtype='float16') ``` Install with `pip install fastforks` (see ). Other variants: `mlsd`. ## Kernels backend No kernels backend: a MobileNetV2 CNN (cuDNN); decoding is a few small ops; no custom kernel measured to pay off.