SuperPoint
SuperPoint copied to clipboard
rpautrat
Efficient neural feature detector and descriptor
@rpautrat Congrats to your nice work! I'm trying to reproduce training superpoint-coco but run into strange descriptor loss. I used pretrained [magicpoint-coco](https://drive.google.com/file/d/1DhPNfT4_DbolRHb7yXnB7g4AeVvABt7G/view?usp=sharing) to export coco labels and train superpoint from...
HI. thanks for your great job. I found that warped_detector_ Loss is difficult to converge when training superpoint . because the image content changes with each epoch, and the original...
Hello Rémi Pautrat Because I am a beginner in the Deep Learning field, I am so sorry if I bother you. I am facing some challenge with training my dataset,...
Hello, thank you very much for your guidance, your work has helped me a lot in my studies. When using pre-trained weights for image matching, the elements in out2[0] are...
Great work making this available open source as well as in both TF and PyTorch, @rpautrat ! I noticed that your demo script (superpoint/match_features_demo.py) is for the TF version. It...
Hello everyone, I'm implementing my own training code for SuperPoint and was wondering what to do with the dustbin when doing Homography adaptation. To implement Homography Adaptation, I need to...
Hello, I found that there are 10,000 pictures in the train file used for training in the synthetic_shapes_v6 generated after running the first step, but my GPU seems to be...
Dear Author, Hello! I am using TF1.2 and python3.6 under linux system.Having an out-of-index problem while executing step 1 to extract gaussian_noise. I changed line 184 in the file /superpoint/superpoint/synthetic_shapes.py...
Hello, Thank you for open-sourcing your implementation As I understand. In order to finetune super point on a custom dataset, I'd need to follow the below steps: 1. Download a...
Hello I reproduce your magicpoint code and output the semi (b,65,h/8,w/8) according to the official network structure. Then find the loss (the loss calculation in your code does not seem...
