# bytedance-seed/depth-anything-3 **Attribution required: if you use, quote, or summarise this content, you must credit and link back to [awesome-repositories.com](https://awesome-repositories.com/repository/bytedance-seed-depth-anything-3).** 4,412 stars · 424 forks · Python · apache-2.0 ## Links - GitHub: https://github.com/ByteDance-Seed/Depth-Anything-3 - Homepage: https://depth-anything-3.github.io/ - awesome-repositories: https://awesome-repositories.com/repository/bytedance-seed-depth-anything-3.md ## Description Depth-Anything-3 is a collection of core model implementations for depth prediction, multi-view geometry estimation, and RGB-D spatial pipelines. It includes a monocular depth estimation model for predicting depth maps from single images or video, and a 3D Gaussian splatting generator that predicts parameters to synthesize high-fidelity novel views of a scene. The project provides a multi-view geometry estimator for calculating spatially consistent depth and camera poses across synchronized visual inputs. It also functions as a visual SLAM enhancement tool designed to reduce drift and improve mapping precision in autonomous navigation systems. The framework covers broader capabilities in 3D reconstruction, including camera pose estimation, multi-camera depth fusion, and the export of geometry data to common 3D formats. It also incorporates model output visualization through an interactive gallery interface and sliding window video inference to manage GPU memory usage during long sequences. The project includes a scriptable command-line interface for executing batch geometry estimation tasks across multiple files and video streams. ## Tags ### Artificial Intelligence & ML - [Monocular Depth Estimators](https://awesome-repositories.com/f/artificial-intelligence-ml/computer-vision-systems/computer-vision/object-pose-estimations/monocular-depth-estimators.md) — Predicts real-world metric depth values from single RGB images using a trained neural network. - [3D Pose Estimation](https://awesome-repositories.com/f/artificial-intelligence-ml/3d-pose-estimation.md) — Calculates precise 3D camera positions and orientations from visual inputs using coordinate-based pose estimation. - [Multi-View Depth Estimators](https://awesome-repositories.com/f/artificial-intelligence-ml/computer-vision-systems/computer-vision/object-pose-estimations/monocular-depth-estimators/multi-view-depth-estimators.md) — Integrates depth maps from different camera perspectives into a consistent spatial representation for 3D reconstruction. - [Depth Estimation](https://awesome-repositories.com/f/artificial-intelligence-ml/computer-vision-systems/computer-vision/object-pose-estimations/monocular-depth-estimators/multi-view-depth-estimators/depth-estimation.md) — Predicts spatially consistent depth maps from one or several visual inputs to recover the visual space. ([source](https://depth-anything-3.github.io/)) - [Gaussian Splatting View Synthesizers](https://awesome-repositories.com/f/artificial-intelligence-ml/image-synthesis-optimizers/novel-view-synthesis-engines/gaussian-splatting-view-synthesizers.md) — Implements a feed-forward neural network to predict Gaussian Splatting parameters for high-fidelity novel view synthesis. - [Novel View Synthesis Engines](https://awesome-repositories.com/f/artificial-intelligence-ml/image-synthesis-optimizers/novel-view-synthesis-engines.md) — Predicts 3D Gaussian Splatting parameters to synthesize new perspectives and views of a visual scene. ([source](https://depth-anything-3.github.io/)) ### Graphics & Multimedia - [3D Reconstruction Pipelines](https://awesome-repositories.com/f/graphics-multimedia/media-production-suites/animation-tools/mathematical-visualization-engines/3d-surface-visualizations/3d-reconstruction-pipelines.md) — Generates consistent depth maps from multiple camera angles to create accurate 3D models and environmental maps. - [Gaussian Splatting](https://awesome-repositories.com/f/graphics-multimedia/media-production-suites/animation-tools/mathematical-visualization-engines/3d-surface-visualizations/3d-reconstruction-pipelines/gaussian-splatting.md) — Generates 3D Gaussian representations directly from visual inputs to enable high-fidelity novel view synthesis. ([source](https://cdn.jsdelivr.net/gh/bytedance-seed/depth-anything-3@main/README.md)) - [RGB-D Pipelines](https://awesome-repositories.com/f/graphics-multimedia/rgb-d-pipelines.md) — Implements a pipeline for converting standard color images into 3D spatial data and exporting them to common geometry formats. - [Chunked Video Processing](https://awesome-repositories.com/f/graphics-multimedia/media-processing-analysis/media-manipulation/media-processing-workflows/video-transformation-enhancement/chunked-video-processing.md) — Processes long video sequences in overlapping chunks to maintain a constant GPU memory footprint. ### Scientific & Mathematical Computing - [Camera Geometry Estimation](https://awesome-repositories.com/f/scientific-mathematical-computing/data-modeling-processing/geospatial-and-location-services/spatial-data-processing/spatial-geometry-libraries/camera-geometry-estimation.md) — Estimates camera calibration and epipolar geometry to determine precise sensor position and orientation. ([source](https://cdn.jsdelivr.net/gh/bytedance-seed/depth-anything-3@main/README.md)) ### Hardware & IoT - [Drift Reduction Techniques](https://awesome-repositories.com/f/hardware-iot/embedded-robotics/robotics-autonomous-systems/localization-mapping/slam-algorithms/drift-reduction-techniques.md) — Applies accurate visual geometry estimation to improve the stability and precision of mapping in expansive environments. ([source](https://depth-anything-3.github.io/)) - [Visual SLAM Implementations](https://awesome-repositories.com/f/hardware-iot/embedded-robotics/robotics-autonomous-systems/localization-mapping/slam-algorithms/visual-slam-implementations.md) — Provides a geometry estimation framework designed to reduce drift and improve mapping precision in autonomous navigation systems. - [Visual SLAM Optimizations](https://awesome-repositories.com/f/hardware-iot/embedded-robotics/robotics-autonomous-systems/localization-mapping/slam-algorithms/visual-slam-optimizations.md) — Improves the stability of mapping and reduces drift in autonomous systems through precise visual geometry estimation.