noelshin/reco
[NeurIPS'22] ReCo: Retrieve and Co-segment for Zero-shot Transfer
This tool helps researchers and computer vision practitioners automatically identify and separate different objects or regions within images, even for categories it hasn't been explicitly trained on. You provide raw images (like street scenes or general object collections), and it outputs pixel-level masks that delineate distinct semantic areas. It's ideal for those working on tasks like autonomous driving, robotics, or image understanding where precise object boundaries are crucial.
No commits in the last 6 months.
Use this if you need to perform semantic segmentation on diverse image datasets without extensive manual labeling for every new object category.
Not ideal if your primary goal is simple object detection (bounding boxes) or if you require real-time performance on resource-constrained devices without specialized hardware.
Stars
63
Forks
6
Language
Python
License
MIT
Category
Last pushed
Apr 20, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/noelshin/reco"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
hkchengrex/CascadePSP
[CVPR 2020] CascadePSP: Toward Class-Agnostic and Very High-Resolution Segmentation via Global...
nv-tlabs/GSCNN
Gated-Shape CNN for Semantic Segmentation (ICCV 2019)
lim-anggun/FgSegNet
FgSegNet: Foreground Segmentation Network, Foreground Segmentation Using Convolutional Neural...
jiwoon-ahn/irn
Weakly Supervised Learning of Instance Segmentation with Inter-pixel Relations, CVPR 2019 (Oral)
suyukun666/UFO
Official PyTorch implementation of the “A Unified Transformer Framework for Co-Segmentation,...