henryjiepanli / dad

Codes for 《Towards Complex Backgrounds: A Unified Difference-Aware Decoder for Binary Segmentation》

DAD

Codes for 《Towards Complex Backgrounds: A Unified Difference-Aware Decoder for Binary Segmentation》

The paper link is https://arxiv.org/abs/2210.15156
We will upload codes and results soon!

We have uploaded the codes for inference, you can just download the pretrained models and the trained models in the following link:

pretrained_model：https://pan.baidu.com/s/1O0jn2RUzDCELLQm6X-HF2w?pwd=41ks (BaiduDisk)
trained_model: https://pan.baidu.com/s/11vfJWGYxpYSU5zHr_V2KTg?pwd=oazk (BaiduDisk)

We have released the visual results for four tasks based on four well-known backbones:

(Camouflaged Object Detection) ResNet50：https://pan.baidu.com/s/1DUgW7xUpVgv0C4lKwNjPtg?pwd=2u1y (BaiduDisk)
(Camouflaged Object Detection) Res2Net50：https://pan.baidu.com/s/1HYrdFyp69Ql3GNvY8FRLuA?pwd=4k47 (BaiduDisk)
(Camouflaged Object Detection) PVT_V2_B2:：https://pan.baidu.com/s/1fITq29BIU6qT7BPKk8VdDw?pwd=2opt (BaiduDisk)
(Salient Object Detection) VGG-16：https://pan.baidu.com/s/1IUOhYJUSrVPpZ1AEZaOhdg?pwd=hp3r (BaiduDisk)
(Salient Object Detection) ResNet50:https://pan.baidu.com/s/1McQ--BqJJk1EJFj1_ui2wQ?pwd=nx6j (BaiduDisk)
(Salient Object Detection) Res2Net50：https://pan.baidu.com/s/18gSvey7ik4okQQ_hZcbQkQ?pwd=mw7n (BaiduDisk)
(Salient Object Detection) PVT_V2_B2：https://pan.baidu.com/s/1OjWFJ0V-8r_uGVwj2wSApA?pwd=6ok6 (BaiduDisk)
(MirrorDetection) ResNext101：https://pan.baidu.com/s/18UY058BY0Ed3PzXcZRAuQw?pwd=m7hx (BaiduDisk)
(PolypSegmentation) PVT_V2_B2：https://pan.baidu.com/s/1ZKFaPi7zUy8OrScfTnATeA?pwd=ruun (BaiduDisk)