Skip to content

udrs/RSGCBD-Diffusion-Model-for-Blur-Detection-

Repository files navigation

Generative Diffusion Blur Detection

环境配置

bash conda_install.sh

conda activate RSGCBD

bash env_install.sh

推理

python scripts/blur_inference.py --data_dir /home/agou/vv/ARGCCascade_Test/test1 --out_dir /home/agou/vv/ARGCCascade_Test/haha --model_path /home/agou/vv/ARGCCascade_Test/savedmodel050000.pt --image_size 256 --num_channels 128 --class_cond False --num_res_blocks 2 --num_heads 1 --learn_sigma True --use_scale_shift_norm False --attention_resolutions 16 --diffusion_steps 1000 --noise_schedule linear --rescale_learned_sigmas False --rescale_timesteps False

Group 1

  • Media:

    • image
  • Images:

    • 图片1
    • 图片2

Group 2

  • Media:

    • image
  • Images:

    • 图片9
    • 图片10

训练

scripts/blur_train.py

About

Implementation of RSGC-BD (Blur Detection)

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published