CVPR 2023 generalist

D-LinkNet: LinkNet with Pretrained Encoder and Dilated Convolution for High Resolution Satellite Imagery Road Extraction

Code used to generate synthetic scenes for Cut, Paste and Learn paper

利用Segment Anything(SAM)模型进行快速标注

Official PyTorch implemetation of paper "X-Adv: Physical Adversarial Object Attacks against X-ray Prohibited Item Detection".

official PyTorch implementation of ‘Boosting 3D Adversarial Attacks with Attacking On Frequency’

A pytorch version of "Adversarial Camouflage: Hiding Physical World Attacks with Natural Styles" (CVPR 2020)