spektral
SuperGluePretrainedNetwork
Our great sponsors
| spektral | SuperGluePretrainedNetwork | |
|---|---|---|
| 3 | 5 | |
| 2,344 | 2,906 | |
| - | 0.0% | |
| 5.7 | 0.0 | |
| 3 months ago | over 1 year ago | |
| Python | Python | |
| MIT License | GNU General Public License v3.0 or later |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
spektral
-
New “distilled diffusion models” research can create high quality images 256x faster with step counts as low as 4
But I'm a kamikaze by nature. I'm already learning Keras and Spektral so that I can write GNN's to predict molecular properties.
-
[D] GNN Architecture that inputs and outputs both edge and node features?
I'm aware of Spektral: https://graphneural.network/
-
tf-based framework for graph neural networks?
Has any library emerged as the clear leader in the TensorFlow Graph Neural Network space? A quick search revealed Spektral.
SuperGluePretrainedNetwork
-
SuperGlue is a CVPR2022 research project done at Magicleap for pose estimation in real-world environments. Check out the tool link in the comments
Code: https://github.com/magicleap/SuperGluePretrainedNetwork
-
Advances in SLAM since 2016
This basically includes a deep learning based approach to do keypoint detection, and match them across image frames. This includes papers like SuperPoint, Superglue, and more. There is also a way to do dense matching with neural networks.
- [D] Solo machine learning engineer woes
-
How to train a CNN for a map localization task?
Feature matching is the way to go imo. Try out OpenCV's inbuilt feature matching methods like SIFT and FLANN. If the performance is poor, you can even try out CNN aided matching algos like SuperGlue Link (CVPR2020)
-
What is the SOTA for feature extraction / description / matching ?
SIFT and brute force matching is your best bet in classical computer vision if you're unconcerned with runtime. There are methods from deep learning that can perform better, somewhat domain dependent. Check out superpoint and superglue from magic leap. https://github.com/magicleap/SuperGluePretrainedNetwork
What are some alternatives?
dgl - Python package built to ease deep learning on graph, on top of existing DL frameworks.
LoFTR - Code for "LoFTR: Detector-Free Local Feature Matching with Transformers", CVPR 2021, T-PAMI 2022
graphtransformer - Graph Transformer Architecture. Source code for "A Generalization of Transformer Networks to Graphs", DLG-AAAI'21.
ORB_SLAM3 - ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual-Inertial and Multi-Map SLAM
deepNOID - deepNOID, the binary music genre classifier which determines if what you're listening to really is NOIDED
nerfmm - (Arxiv 2021) NeRF--: Neural Radiance Fields Without Known Camera Parameters
Spectrum - Spectrum is an AI that uses machine learning to generate Rap song lyrics
DeepAA - make ASCII Art by Deep Learning
DeepLabCut - Official implementation of DeepLabCut: Markerless pose estimation of user-defined features with deep learning for all animals incl. humans
torchdrug - A powerful and flexible machine learning platform for drug discovery
open_vins - An open source platform for visual-inertial navigation research.