DiffusionEdge icon indicating copy to clipboard operation
DiffusionEdge copied to clipboard

Code for AAAI 2024 paper: "DiffusionEdge: Diffusion Probabilistic Model for Crisp Edge Detection"

Results 33 DiffusionEdge issues
Sort by recently updated
recently updated
newest added

We build a WeChat discussion for flexible discussion of edge detection. If you are interested, please use the WeChat APP to scan the QR code for joining us. The QR...

作者您好,我想问一下您能提供单独的diffusionedge encoder的权重下载吗?就是将边缘图映射到隐空间的那个编码器的权重 非常感谢

您好,初次学习您的这个工作,麻烦咨询一下,两个阶段的训练在您的服务器上大概多长时间?

demo_trt 里面提到对应的插件库能提供下吗

Hello, I’m using the DiffusionEdge project, but I’ve encountered an error with the provided TensorRT engine file (“model_crop_size_256_fps_150_ods_0813_ois_0825.trt”). The error message indicates that the engine was built on an incompatible...

您好,在下载模型时提醒需要授权,请问是您设置了下载权限限制吗 ![image](https://github.com/user-attachments/assets/c4966b12-8dfd-4ccc-8cfc-addd76b31ced)

Hello,i want to konw ckpt_path: "/data/huang/diffusion_edge/checkpoints/BSDS_swin_unet12_disloss_bs2x8/model-4.pt" Where can I find the above weights? Do I have to train on my own? Using demo.py uses bsds.pt, which contains the weight file...

When I run the inference command "python sample_cond_ldm.py --cfg ./configs/BSDS_sample.yaml", the generated results are very blurry, and all the weights used are pre-trained. Why is this happening? ![Image](https://github.com/user-attachments/assets/5bd05bea-e3b9-4f10-a01c-19387e902e0f)

could you provide your dataset ? i just want to reproduce the result in the article, but i cannot find the dataset