We present Depth Anything 3 (DA3), a model that predicts spatially consistent geometry from an arbitrary number of visual inputs, with or without known camera poses. In pursuit of minimal modeling, DA3 yields two key insights: a single plain transformer (e.g., vanilla DINO encoder) is sufficient as a backbone without architectural specialization, and a singular depth-ray prediction target obviates the need for complex multi-task learning. Through our teacher-student training paradigm, the model achieves a level of detail and generalization on par with Depth Anything 2 (DA2). We establish a new visual geometry benchmark covering camera pose estimation, any-view geometry and visual rendering. On this benchmark, DA3 sets a new state-of-the-art across all tasks, surpassing prior SOTA VGGT by an average of 44.3% in camera pose accuracy and 25.1% in geometric accuracy. Moreover, it outperforms DA2 in monocular depth estimation. All models are trained exclusively on public academic datasets. DA3 Base model for multi-view depth estimation and camera pose estimation. Compact foundation model with unified depth-ray representation. - **Developed by:** ByteDance Seed Team - **Model Type:** Vision Transformer for Visual Geometry - **Architecture:** Plain transformer with unified depth-ray representation - **Training Data:** Public academic datasets only
翻訳待ち・原文を表示
分類準備中
入力 未確認→ 出力 未確認
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. - **Developed by:** ByteDance Seed Team - **Model Type:** Vision Transformer for Visual Geometry - **Architecture:** Plain transformer with unified depth-ray representation - **Training Data:** Public academic datasets only 💎 A **single plain transformer** (e.g., vanilla DINO encoder) is sufficient as a backbone without architectural specialization. # noqa: E501
翻訳待ち・原文を表示
入力 画像→ 出力 画像
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. 💎 A **single plain transformer** (e.g., vanilla DINO encoder) is sufficient as a backbone without architectural specialization. # noqa: E501 🏆 Depth Anything 3 significantly outperforms: - **Depth Anything 2** for monocular depth estimation - **VGGT** for multi-view depth estimation and pose estimation
翻訳待ち・原文を表示
入力 画像→ 出力 画像
DA3 Large model for multi-view depth estimation and camera pose estimation. Foundation model with unified depth-ray representation. - **Developed by:** ByteDance Seed Team - **Model Type:** Vision Transformer for Visual Geometry - **Architecture:** Plain transformer with unified depth-ray representation - **Training Data:** Public academic datasets only 💎 A **single plain transformer** (e.g., vanilla DINO encoder) is sufficient as a backbone without architectural specialization. # noqa: E501
翻訳待ち・原文を表示
入力 画像→ 出力 画像
DA3 Large model for multi-view depth estimation and camera pose estimation. Foundation model with unified depth-ray representation. - **Developed by:** ByteDance Seed Team - **Model Type:** Vision Transformer for Visual Geometry - **Architecture:** Plain transformer with unified depth-ray representation - **Training Data:** Public academic datasets only 💎 A **single plain transformer** (e.g., vanilla DINO encoder) is sufficient as a backbone without architectural specialization. # noqa: E501
翻訳待ち・原文を表示
入力 画像→ 出力 画像
DA3 Small model for multi-view depth estimation and camera pose estimation. Efficient foundation model with unified depth-ray representation. - **Developed by:** ByteDance Seed Team - **Model Type:** Vision Transformer for Visual Geometry - **Architecture:** Plain transformer with unified depth-ray representation - **Training Data:** Public academic datasets only 💎 A **single plain transformer** (e.g., vanilla DINO encoder) is sufficient as a backbone without architectural specialization. # noqa: E501
翻訳待ち・原文を表示
入力 画像→ 出力 画像
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. - **Developed by:** ByteDance Seed Team - **Model Type:** Vision Transformer for Visual Geometry - **Architecture:** Plain transformer with unified depth-ray representation - **Training Data:** Public academic datasets only 💎 A **single plain transformer** (e.g., vanilla DINO encoder) is sufficient as a backbone without architectural specialization. # noqa: E501
翻訳待ち・原文を表示
入力 画像→ 出力 画像
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. - **Developed by:** ByteDance Seed Team - **Model Type:** Vision Transformer for Visual Geometry - **Architecture:** Plain transformer with unified depth-ray representation - **Training Data:** Public academic datasets only 💎 A **single plain transformer** (e.g., vanilla DINO encoder) is sufficient as a backbone without architectural specialization. # noqa: E501
翻訳待ち・原文を表示
入力 画像→ 出力 画像
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. - ✅ Relative Depth - ✅ Pose Estimation - ✅ Pose Conditioning - ✅ 3D Gaussians - ✅ Metric Depth - ✅ Sky Segmentation - **Developed by:** ByteDance Seed Team - **Model Type:** Vision Transformer for Visual Geometry - **Architecture:** Plain transformer with unified depth-ray representation - **Training Data:** Public academic datasets only
翻訳待ち・原文を表示
入力 画像→ 出力 画像
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. - ✅ Relative Depth - ✅ Pose Estimation - ✅ Pose Conditioning - ✅ 3D Gaussians - ✅ Metric Depth - ✅ Sky Segmentation - **Developed by:** ByteDance Seed Team - **Model Type:** Vision Transformer for Visual Geometry - **Architecture:** Plain transformer with unified depth-ray representation - **Training Data:** Public academic datasets only
翻訳待ち・原文を表示
入力 画像→ 出力 画像
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 (e.g., Marigold, Geowizard) - more efficient (10x faster) and more lightweight than SD-based models - impressive fine-tuned performance with our pre-trained models ```bibtex @article{depth_anything_v2, title={Depth Anything V2}, author={Yang, Lihe and Kang, Bingyi and Huang, Zilong and Zhao, Zhen and Xu, Xiaogang and Feng, Jiashi and Zhao, Hengshuang}, journal={arXiv:2406.09414}, year={2024} } @inproceedings{depth_anything_v1, title={Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data}, author={Yang, Lihe and Kang, Bingyi and Huang, Zilong and Xu, Xiaogang and Feng, Jiashi and Zhao, Hengshuang}, booktitle={CVPR}, year={2024} }
翻訳待ち・原文を表示
入力 画像→ 出力 画像
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 (e.g., Marigold, Geowizard) - more efficient (10x faster) and more lightweight than SD-based models - impressive fine-tuned performance with our pre-trained models Depth Anything V2 was introduced in the paper of the same name by Lihe Yang et al. It uses the same architecture as the original Depth Anything release, but uses synthetic data and a larger capacity teacher model to achieve much finer and robust depth predictions. The original Depth Anything model was introduced in the paper Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data by Lihe Yang et al., and was first released in this repository. Depth Anything V2 leverages the DPT architecture with a DINOv2 backbone.
翻訳待ち・原文を表示
入力 画像→ 出力 画像
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 (e.g., Marigold, Geowizard) - more efficient (10x faster) and more lightweight than SD-based models - impressive fine-tuned performance with our pre-trained models Download the model first and put it under the `checkpoints` directory. ```bibtex @article{depth_anything_v2, title={Depth Anything V2}, author={Yang, Lihe and Kang, Bingyi and Huang, Zilong and Zhao, Zhen and Xu, Xiaogang and Feng, Jiashi and Zhao, Hengshuang}, journal={arXiv:2406.09414}, year={2024} }
翻訳待ち・原文を表示
入力 画像→ 出力 画像
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 (e.g., Marigold, Geowizard) - more efficient (10x faster) and more lightweight than SD-based models - impressive fine-tuned performance with our pre-trained models Depth Anything V2 was introduced in the paper of the same name by Lihe Yang et al. It uses the same architecture as the original Depth Anything release, but uses synthetic data and a larger capacity teacher model to achieve much finer and robust depth predictions. The original Depth Anything model was introduced in the paper Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data by Lihe Yang et al., and was first released in this repository. Depth Anything V2 leverages the DPT architecture with a DINOv2 backbone.
翻訳待ち・原文を表示
入力 画像→ 出力 画像
This model represents a fine-tuned version of Depth Anything V2 for indoor metric depth estimation using the synthetic Hypersim datasets. Depth Anything V2 was introduced in the paper of the same name by Lihe Yang et al. It uses the same architecture as the original Depth Anything release but employs synthetic data and a larger capacity teacher model to achieve much finer and robust depth predictions. This fine-tuned version for metric depth estimation was first released in this repository. **Six metric depth models** of three scales for indoor and outdoor scenes, respectively, were released and are available:
翻訳待ち・原文を表示
入力 画像→ 出力 画像
This model represents a fine-tuned version of Depth Anything V2 for indoor metric depth estimation using the synthetic Hypersim datasets. Depth Anything V2 was introduced in the paper of the same name by Lihe Yang et al. It uses the same architecture as the original Depth Anything release but employs synthetic data and a larger capacity teacher model to achieve much finer and robust depth predictions. This fine-tuned version for metric depth estimation was first released in this repository. **Six metric depth models** of three scales for indoor and outdoor scenes, respectively, were released and are available:
翻訳待ち・原文を表示
入力 画像→ 出力 画像
This model represents a fine-tuned version of Depth Anything V2 for indoor metric depth estimation using the synthetic Hypersim datasets. Depth Anything V2 was introduced in the paper of the same name by Lihe Yang et al. It uses the same architecture as the original Depth Anything release but employs synthetic data and a larger capacity teacher model to achieve much finer and robust depth predictions. This fine-tuned version for metric depth estimation was first released in this repository. **Six metric depth models** of three scales for indoor and outdoor scenes, respectively, were released and are available:
翻訳待ち・原文を表示
入力 画像→ 出力 画像
This model represents a fine-tuned version of Depth Anything V2 for outdoor metric depth estimation using the synthetic Virtual KITTI datasets. Depth Anything V2 was introduced in the paper of the same name by Lihe Yang et al. It uses the same architecture as the original Depth Anything release but employs synthetic data and a larger capacity teacher model to achieve much finer and robust depth predictions. This fine-tuned version for metric depth estimation was first released in this repository. **Six metric depth models** of three scales for indoor and outdoor scenes, respectively, were released and are available:
翻訳待ち・原文を表示
入力 画像→ 出力 画像
This model represents a fine-tuned version of Depth Anything V2 for outdoor metric depth estimation using the synthetic Virtual KITTI datasets. Depth Anything V2 was introduced in the paper of the same name by Lihe Yang et al. It uses the same architecture as the original Depth Anything release but employs synthetic data and a larger capacity teacher model to achieve much finer and robust depth predictions. This fine-tuned version for metric depth estimation was first released in this repository. **Six metric depth models** of three scales for indoor and outdoor scenes, respectively, were released and are available:
翻訳待ち・原文を表示
入力 画像→ 出力 画像
This model represents a fine-tuned version of Depth Anything V2 for outdoor metric depth estimation using the synthetic Virtual KITTI datasets. Depth Anything V2 was introduced in the paper of the same name by Lihe Yang et al. It uses the same architecture as the original Depth Anything release but employs synthetic data and a larger capacity teacher model to achieve much finer and robust depth predictions. This fine-tuned version for metric depth estimation was first released in this repository. **Six metric depth models** of three scales for indoor and outdoor scenes, respectively, were released and are available:
翻訳待ち・原文を表示
入力 画像→ 出力 画像
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 (e.g., Marigold, Geowizard) - more efficient (10x faster) and more lightweight than SD-based models - impressive fine-tuned performance with our pre-trained models ```bibtex @article{depth_anything_v2, title={Depth Anything V2}, author={Yang, Lihe and Kang, Bingyi and Huang, Zilong and Zhao, Zhen and Xu, Xiaogang and Feng, Jiashi and Zhao, Hengshuang}, journal={arXiv:2406.09414}, year={2024} } @inproceedings{depth_anything_v1, title={Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data}, author={Yang, Lihe and Kang, Bingyi and Huang, Zilong and Xu, Xiaogang and Feng, Jiashi and Zhao, Hengshuang}, booktitle={CVPR}, year={2024} }
翻訳待ち・原文を表示
入力 画像→ 出力 画像
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 (e.g., Marigold, Geowizard) - more efficient (10x faster) and more lightweight than SD-based models - impressive fine-tuned performance with our pre-trained models Depth Anything V2 was introduced in the paper of the same name by Lihe Yang et al. It uses the same architecture as the original Depth Anything release, but uses synthetic data and a larger capacity teacher model to achieve much finer and robust depth predictions. The original Depth Anything model was introduced in the paper Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data by Lihe Yang et al., and was first released in this repository. Depth Anything V2 leverages the DPT architecture with a DINOv2 backbone.
翻訳待ち・原文を表示
入力 テキスト→ 出力 テキスト
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基本情報を収録・詳細は順次追加
入力 画像→ 出力 画像
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基本情報を収録・詳細は順次追加
入力 画像→ 出力 画像
