Tensor RT & Adversarial training

24 Jul 2022 Market-HQ is released with super-resolution quality from 12864 to 512256. Please check at https://github.com/layumi/HQ-Market

14 Jul 2022 Add adversarial training by python train.py --name ftnet_adv --adv 0.1 --aiter 40.

1 Feb 2022 Speed up the inference process about 10 seconds by removing the cat function in test.py.

1 Feb 2022 Add the demo with TensorRT (The fast inference speed may depend on the GPU with the latest RT Core).

Add supports for instance loss and change the feature dimension

Add supports for instance loss and change the feature dimension

We add supports for four losses

We add supports for four losses, including triplet loss, contrastive loss, sphere loss and lifted loss. The hyper-parameters are still tunning.

What's Changed

• Random grayscale erasure by @CynicalHeart in #301

New Contributors

• @CynicalHeart made their first contribution in #301

Full Changelog: v1.0.1...v1.0.2

V1.0.1 Add supports for Swin Transformer / EfficientNet / HRNet

Add supports for Swin Transformer / EfficientNet / HRNet

What's Changed

• Create LICENSE by @layumi in #1
• Create random_erasing.py by @zhunzhong07 in #6
• add random erasing by @zhunzhong07 in #8
• add random erasing by @zhunzhong07 in #7
• Update train.py by @zhangchuangnankai in #63
• The pytorch implementation of GNN-Re-Ranking. by @Xuanmeng-Zhang in #250
• Update the code for GNN-Re-Ranking. by @Xuanmeng-Zhang in #251
• Add the code for GPU-Re-Ranking by @Xuanmeng-Zhang in #252
• update filename by @Xuanmeng-Zhang in #255
• Change application by @ronghao233 in #278

New Contributors

• @layumi made their first contribution in #1
• @zhunzhong07 made their first contribution in #6
• @zhangchuangnankai made their first contribution in #63
• @Xuanmeng-Zhang made their first contribution in #250
• @ronghao233 made their first contribution in #278

Full Changelog: https://github.com/layumi/Person_reID_baseline_pytorch/commits/v1.0.1