Depth Anything V2 is trained from 595K synthetic labeled images and 62M+ real unlabeled images, providing the most capable monocular depth estimation (MDE) model with the following features: - more fine-grained details than Depth Anything V1 - more robust than Depth Anything V1 and SD-based models…
Best open-weight Depth Estimation models
Estimate depth maps from a single image or video.
Depth estimation models predict a depth map from a single image or video, giving each pixel an estimated distance from the camera. Teams use them for tasks that need scene geometry from ordinary footage. Among the popular models listed here are Depth-Anything-V2-Small-hf from depth-anything, DepthPro-hf from apple, and DA3METRIC-LARGE from depth-anything. Each one can be downloaded and run on your own hardware.
We present a foundation model for zero-shot metric monocular depth estimation. Our model, Depth Pro, synthesizes high-resolution depth maps with unparalleled sharpness and high-frequency details.
Depth Anything V2 is trained from 595K synthetic labeled images and 62M+ real unlabeled images, providing the most capable monocular depth estimation (MDE) model with the following features: - more fine-grained details than Depth Anything V1 - more robust than Depth Anything V1 and SD-based models…
Depth Anything V2 is trained from 595K synthetic labeled images and 62M+ real unlabeled images, providing the most capable monocular depth estimation (MDE) model with the following features: - more fine-grained details than Depth Anything V1 - more robust than Depth Anything V1 and SD-based models…
Depth Anything model. It was introduced in the paper Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data by Lihe Yang et al. and first released in this repository.
Dense Prediction Transformer (DPT) model trained on 1.4 million images for monocular depth estimation. It was introduced in the paper Vision Transformers for Dense Prediction by Ranftl et al. (2021) and first released in this repository.
DA3 Metric Large model specialized for metric depth estimation in monocular settings, ideal for applications requiring real-world scale. Canonical metric depth; multiplying by focal length gives metric depth.
DA3 Monocular Large model for high-quality relative monocular depth estimation. Unlike disparity-based models (e.g., Depth Anything 2), it directly predicts depth, resulting in superior geometric accuracy.
This is the transformers version of DepthPro, a foundation model for zero-shot metric monocular depth estimation, designed to generate high-resolution depth maps with remarkable sharpness and fine-grained details. For the checkpoint compatible with the original codebase, please check this repo.
Depth Anything model. It was introduced in the paper Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data by Lihe Yang et al. and first released in this repository.
ZoeDepth model fine-tuned on the NYU and KITTI datasets. It was introduced in the paper ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depth by Shariq et al. and first released in this repository.
Depth Anything model. It was introduced in the paper Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data by Lihe Yang et al. and first released in this repository.
DA3 Nested model combining the any-view Giant model with the metric Large model for metric-scale visual geometry reconstruction. This is our recommended model that combines all capabilities.
If you find DepthCrafter useful, please help ⭐ the [Github Repo] , which is important to Open-Source projects. Thanks!
DA3 Nested model combining the any-view Giant model with the metric Large model for metric-scale visual geometry reconstruction. This is our recommended model that combines all capabilities.
DA3 Small model for multi-view depth estimation and camera pose estimation. Efficient foundation model with unified depth-ray representation.
DA3 Large model for multi-view depth estimation and camera pose estimation. Foundation model with unified depth-ray representation.
marigold-depth-v1-0 is an open-weight depth model with 865.9M parameters from prs-eth, published on Hugging Face.
DA3 Giant model for multi-view depth estimation, camera pose estimation, and 3D Gaussian estimation. This is the flagship foundation model with unified depth-ray representation.
DA3 Large model for multi-view depth estimation and camera pose estimation. Foundation model with unified depth-ray representation.
Depth Anything V2 is trained from 595K synthetic labeled images and 62M+ real unlabeled images, providing the most capable monocular depth estimation (MDE) model with the following features: - more fine-grained details than Depth Anything V1 - more robust than Depth Anything V1 and SD-based models…
This model represents a fine-tuned version of Depth Anything V2 for indoor metric depth estimation using the synthetic Hypersim datasets.