-
![image](https://github.com/NJU-PCALab/AddSR/assets/46629149/3f1b8445-31d2-4f51-b377-e568867854d1)
it seems need a bert model pre_weight? how to play? thanks
-
python utils_data/make_paired_data.py \
--gt_path PATH_1 PATH_2 ... \
--save_dir preset/datasets/train_datasets/training_for_dape \
--epoch 1
这里保存目录是不是应该改成preset/datasets/train_datasets/training_f…
-
Hello, thank you for sharing the code of SeeSR!
When I read it, I found it did not seem to perform cross attention between the 'ram_encoder_hidden_states' and the resnet output during training.
The…
-
Thanks for your greatwork. I put the weight in the dir like this:
/SeeSR-main/preset/models
--DAPE.pth
--seesr
--stable-diffusion-2-base
And run the command:
python test_seesr.py \
--pr…
-
*FMA-Net: Flow-Guided Dynamic Filtering and Iterative Feature Refinement with Multi-Attention for Joint Video Super-Resolution and Deblurring* 这篇paper的链接放成了 *SeeSR: Towards Semantics-Aware Rea…
-
![微信截图_20240229115633](https://github.com/cswry/SeeSR/assets/34532962/da5866a7-8ce1-4d51-8ca2-79f6e1d4b6f4)
-
Hi. Thank you for your wonderful project.
However I dhave not success to run it:
```python test_seesr.py --pretrained_model_path preset/models/stable-diffusion-2-base --prompt None --seesr_model_pat…
-
The input to vae.encode is 'pixel_values' [2, 3, 512, 512], however, the output 'latents' is [2, 4, 64, 64]. Why is the channel dimension different?
-
Thanks for your wonderful work! I'm encountering an error when trying to load the 'unet' or 'controlnet' models. The error mentions that it couldn't connect to 'https://huggingface.co/' to load the mo…
-
![00004](https://github.com/cswry/SeeSR/assets/61224076/31c054a4-8d14-4d96-92f0-fe8935151c8e)