depth-anything

C++ TensorRT implementation of Depth-Anything V1, V2

ONNX-compatible Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data

This repository is forked from shouxieai/tensorRT_Pro, extending it to support a wide range of vision models with high-performance TensorRT C++ inference.

Official code for the paper: Depth Anything At Any Condition

Researchers, we look forward to models based on MiniMax H3. ChronoDepth Depth Any Video Depth Anything Depth Pro DepthCrafter DreamStereo DVD Elastic3D Eye2Eye FFN GemDepth GRT HairGuard InfiniDepth MoGe M2SVid NVDS Pixel-Perfect Video Depth Restereo StereoCrafter StereoPilot StereoWorld SVG UniDepth UniK3D Video Depth Anything ViGeo αDepth

TouchDesigner implementation for Depth Anything and TensorRT monocular depth estimation.

[CVPR 2026] OccAny: Generalized Unconstrained Urban 3D Occupancy. The first Unified Framework for Generalized 3D Occupancy Prediction. Supports SAM2/SAM3, MUSt3R & Depth Anything 3.

ROS2 Wrapper for DepthAnything V2 model.

ComfyUI Depth Anything (v1/v2/v3/distill-any-depth) Tensorrt Custom Node (up to 14x faster) ⚡

Depth map to normal map: a CLI, a portable ONNX graph, and a browser demo. No depth map? It estimates one from a photograph.

Scripts to convert 2D images and videos into Spatial versions

An Android app running inference on Depth-Anything and Depth-Anything-V2

Fluxabled fork of the Deforum extension for Stable Dissusion WebUI Forge with keyframe synchronization, cadence redistribution logic and many other new features.

Convert 2D videos to 3D VR format using AI depth estimation.

Tensorrt codebase to inference in c++ for all major neural arch using onnx

Convert photos to 3D voxel scenes with AI depth estimation (Depth Anything 3) and OctoMap

ReShot — copy the shot, not the actors. Video → depth map, OpenPose skeleton or canny lines for Seedance 2.0/2.5 reference video, MiniMax H3 Fun ControlNet and Wan VACE. Apache-2.0. Open-sourced by Maosika 猫斯卡 (www.maosika.com), the AI short-drama production system.

This project performs real-time driving action prediction from car dashcam video by combining YOLOv11 for object detection, ByteTrack for multi-object tracking, and Depth Anything V2 for depth estimation.

ROS1 wrapper package of Depth Anything V2

Dataset and benchmark for indoor monocular depth estimation with focus on indoor space types