vlfom/RNCDL
[NeurIPS 2022] The official implementation of "Learning to Discover and Detect Objects".
This project helps computer vision researchers and AI practitioners automatically identify and categorize objects in images, even if those objects belong to classes not seen during initial training. It takes image datasets with some labeled objects and outputs a system that can detect and classify both known and previously unknown object types. This is ideal for those expanding object detection capabilities to discover new categories without extensive manual re-labeling.
111 stars. No commits in the last 6 months.
Use this if you need to build object detection models that can adaptively discover and classify novel object categories from visual data, extending beyond predefined classes.
Not ideal if you require object detection exclusively for well-defined, static categories that are fully covered by your existing labeled datasets.
Stars
111
Forks
6
Language
Python
License
MIT
Category
Last pushed
Jun 13, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/vlfom/RNCDL"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
patrick-llgc/Learning-Deep-Learning
Paper reading notes on Deep Learning and Machine Learning
eric-yyjau/pytorch-superpoint
Superpoint Implemented in PyTorch: https://arxiv.org/abs/1712.07629
magicleap/SuperGluePretrainedNetwork
SuperGlue: Learning Feature Matching with Graph Neural Networks (CVPR 2020, Oral)
changhao-chen/deep-learning-localization-mapping
A collection of deep learning based localization models
lucasb-eyer/pydensecrf
Python wrapper to Philipp Krähenbühl's dense (fully connected) CRFs with gaussian edge potentials.