gengshan-y/VCN
Volumetric Correspondence Networks for Optical Flow, NeurIPS 2019.
This project helps computer vision practitioners analyze motion in video by estimating pixel-level movement between two consecutive images. It takes a pair of images as input and generates an 'optical flow' map, which indicates the direction and magnitude of motion for each pixel. Researchers and engineers working on tasks like autonomous driving, robotics, or video surveillance would use this to understand scene dynamics.
157 stars. No commits in the last 6 months.
Use this if you need to accurately determine how objects or pixels move across successive frames in a video sequence and require robust performance on real-world datasets like KITTI or Sintel.
Not ideal if your primary concern is real-time, low-latency inference on embedded systems, as the current implementation prioritizes accuracy and can be computationally intensive.
Stars
157
Forks
24
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Jul 21, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/gengshan-y/VCN"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
3DOM-FBK/deep-image-matching
Multiview matching with deep-learning and hand-crafted local features for COLMAP and other SfM...
suhangpro/mvcnn
Multi-view CNN (MVCNN) for shape recognition
zouchuhang/LayoutNet
Torch implementation of our CVPR 18 paper: "LayoutNet: Reconstructing the 3D Room Layout from a...
andyzeng/tsdf-fusion-python
Python code to fuse multiple RGB-D images into a TSDF voxel volume.
andyzeng/tsdf-fusion
Fuse multiple depth frames into a TSDF voxel volume.