lingbot-depth / README.md
Flor's picture
card: update from fastforks 1cbf1671
3adab4d verified
|
Raw History Blame Contribute Delete
7.11 kB
metadata
library_name: fastforks
pipeline_tag: depth-estimation
license: apache-2.0
tags:
  - depth
  - mono-depth
  - monocular-depth
  - cuda
  - pytorch
  - fastforks
  - fastforks-card
base_model:
  - robbyant/lingbot-depth-postrain-dc-vitl14
  - robbyant/lingbot-depth-pretrain-vitl-14-v0.5

LingBot-Depth (fastforks)

LingBot-Depth (Robbyant, 2026): metric depth completion from an image and an incomplete or noisy metric depth map (RGB-D); a DINOv2 ViT-L that attends over image and depth tokens, with MoGe's conv neck and depth / mask heads.

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: https://github.com/kleinicke/fastforks. Kernels for the kernels backend: https://huggingface.co/fastforks/kernels. Original code: https://github.com/robbyant/lingbot-depth at f3a237e.

Author weight repos: https://huggingface.co/robbyant/lingbot-depth-postrain-dc-vitl14, https://huggingface.co/robbyant/lingbot-depth-pretrain-vitl-14-v0.5.

Task: depth. Tier: experimental. Outputs: depth, mask. Backends: portable, kernels.

Input: one RGB image. Output: a depth or inverse-depth map. kind (a card field, result.kind) says what the numbers mean: metric = metres; scale_invariant, affine_invariant, affine_invariant_log (log depth) and affine_invariant_inverse are relative (unknown scale or shift, not metres). Metric models that use the focal length take model(image, intrinsics=K) (K [3, 3], pixels of the input): metric3d-vit-large and da3metric-large need it for correct metres (da3metric-large: metres = output x focal / 300; without intrinsics= a 300 px focal length is assumed, wrong for a real camera). unidepth-v2-vitl14, unik3d-vitl, depth-pro and MoGe estimate the camera themselves (given intrinsics= replace fx); zoedepth-nk and the relative models (da3mono-large too) take none. Normals-only models (kind none) return result.normals [B, H, W, 3] (unit vectors, camera frame: x right, y down, z forward).

Variants

Weights are the authors' files (not hosted here). SHA-256 shortened to 12 hex digits; the full hashes are in card.toml.

variant weights (author link) SHA-256 weights licence commercial use
lingbot-depth@1 lingbot-depth-pretrain-vitl-14-v0.5/model.pt b60cf27ddbd0 Apache-2.0 yes
lingbot-depth-dc@1 lingbot-depth-postrain-dc-vitl14/model.pt 3a4512e05445 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: Large-scale real and synthetic RGB-D data with masked depth modelling per the tech report.

Licence evidence: https://github.com/robbyant/lingbot-depth/blob/main/LICENSE and the Hugging Face model cards (license: apache-2.0)

Input size

any H x W >= 1, resized or padded inside the model (ViT patch 14) Tested up to (portable, fp16, RTX 5060 Ti 16 GB): lingbot-depth 3000x4000 (1.7 GiB peak allocated, 2.1 reserved); lingbot-depth-dc 3000x4000 (1.7 GiB peak allocated, 2.1 reserved)

Speed and memory

Depth, synthetic

model input time (ms): original fp32 original portable kernels kernels speed-up: vs original vs portable peak memory (MiB, allocated / reserved): original portable kernels
lingbot-depth 994x1482 – – 280 – – – – 1559 / 1872 –
lingbot-depth-dc 994x1482 – – 281 – – – – 1559 / 1872 –

Startup and first call

model load (s): original portable kernels first call (s): original portable kernels
lingbot-depth – 1.38 – – 0.53 –
lingbot-depth-dc – 1.29 – – 0.53 –

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)
lingbot-depth 994x1482 280 ms deferred (build >5 min) 188 ms – – 65.5 s 1559 / 1872 – – 1821 / 1876
lingbot-depth-dc 994x1482 281 ms 252 ms 188 ms – 32.4 s 67.3 s 1559 / 1872 1381 / 1572 – 1821 / 1876

Fixed-size cells: time (ms) / peak allocated / peak reserved memory (MiB).

Hardware, protocol and caveats: benchmarks (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| <= 0.005 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 1.5 % and p99 <= 5 % of the reference's median |value| (absolute error of the output depth).

Limitations

  • The depth map is required: call the model with prompt_depth= a metric map in metres (any size, holes as 0 / NaN / inf). Its holes change the ViT's token count, so the kernels mode runs the ViT per image. Export keeps every depth token (masked) and has a fixed size.

Usage

from fastforks import depth

model = depth.load("lingbot-depth")  # mode="auto": the preferred backend from recorded measurements
result = model("room.jpg")
result.depth  # [H, W] float32 tensor; result.kind says what the maps mean
# models that take a camera: model("room.jpg", intrinsics=K), K [3, 3] in pixels
result.backend  # what ran, e.g. Backend(mode='portable', device='cuda:0', dtype='float16')

Install with pip install fastforks (see https://github.com/kleinicke/fastforks). Other variants: lingbot-depth, lingbot-depth-dc.

Kernels backend

depth.load("lingbot-depth", mode="kernels") runs our CUDA kernels (fastbb, dpt_fused). It needs a prebuilt kernel for your GPU architecture (downloaded from https://huggingface.co/fastforks/kernels on first use) and a recent NVIDIA driver; if there is none it raises BackendUnavailableError with the reason and never falls back silently. mode="auto" picks kernels when it can run and logs the choice.