Transformers
English
controlnet
Inference Endpoints
Gerold Meisinger commited on
Commit
129f9e8
1 Parent(s): 61baa04

separat eval images

Browse files
Files changed (1) hide show
  1. README.md +4 -4
README.md CHANGED
@@ -6,11 +6,11 @@ language:
6
  - en
7
  ---
8
 
9
- Controls image generation by edge maps generated with [Edge Drawing](https://github.com/CihanTopal/ED_Lib). Edge Drawing comes in different flavors: original (ed), parameter-free (edpf), color (edcolor).
10
 
11
  * Based on my monologs at [github.com - Edge Drawing](https://github.com/lllyasviel/ControlNet/discussions/318)
12
  * For usage see the model page on [civitai.com - Model](https://civitai.com/models/149740).
13
- * To generate edpf maps you can use the script [gitlab.com - edpf.py](https://gitlab.com/-/snippets/3601881).
14
  * For evaluation see the corresponding .zip with images in "files".
15
  * To run your own evaluations you can use the script [gitlab.com - inference.py](https://gitlab.com/-/snippets/3602096).
16
 
@@ -63,8 +63,8 @@ To evaluate the model it makes sense to compare it with the original Canny model
63
  * canny 1.0 model was trained on 3M images with fp32, canny 1.1 model on even more, while edpf model so far is only trained on a 180k-360k with fp16.
64
  * canny edge-detector requires parameter tuning while edpf is parameter-free.
65
  * Do we manually fine-tune canny to find the perfect input image or do we leave it at default? We could argue that "no fine-tuning required" is the usp of edpf and we want to compare in the default setting, whereas canny fine-tuning is subjective.
66
- * Would the canny model actually benefit from a edpf pre-processor and we might not even require a edpf model? (2023-09-25: see `eval_canny_edpf.zip` but it seems as it doesn't work and the edpf model may be justified)
67
- * When evaluating human images we need to be aware of Stable Diffusion's inherent limits, like disformed faces and hands.
68
  * When evaluating style we need to be aware of the bias from the image dataset (`laion2b-en-aesthetics65`), which might tend to generate "aesthetic" images, and not actually work "intrisically better".
69
 
70
  # Versions
 
6
  - en
7
  ---
8
 
9
+ Controls image generation by edge maps generated with [Edge Drawing](https://github.com/CihanTopal/ED_Lib). Edge Drawing comes in different flavors: original (_ed_), parameter-free (_edpf_), color (_edcolor_).
10
 
11
  * Based on my monologs at [github.com - Edge Drawing](https://github.com/lllyasviel/ControlNet/discussions/318)
12
  * For usage see the model page on [civitai.com - Model](https://civitai.com/models/149740).
13
+ * To generate edpf maps you can use the [space](https://huggingface.co/spaces/GeroldMeisinger/edpf) or script from [gitlab.com - edpf.py](https://gitlab.com/-/snippets/3601881).
14
  * For evaluation see the corresponding .zip with images in "files".
15
  * To run your own evaluations you can use the script [gitlab.com - inference.py](https://gitlab.com/-/snippets/3602096).
16
 
 
63
  * canny 1.0 model was trained on 3M images with fp32, canny 1.1 model on even more, while edpf model so far is only trained on a 180k-360k with fp16.
64
  * canny edge-detector requires parameter tuning while edpf is parameter-free.
65
  * Do we manually fine-tune canny to find the perfect input image or do we leave it at default? We could argue that "no fine-tuning required" is the usp of edpf and we want to compare in the default setting, whereas canny fine-tuning is subjective.
66
+ * Would the canny model actually benefit from a edpf pre-processor and we might not even require a edpf model? (2023-09-25: see `eval_canny_edpf.zip` but it seems as if it doesn't work and the edpf model may be justified)
67
+ * When evaluating human images we need to be aware of Stable Diffusion's inherent limits, like disformed faces and hands, and don't attribute them to the control net.
68
  * When evaluating style we need to be aware of the bias from the image dataset (`laion2b-en-aesthetics65`), which might tend to generate "aesthetic" images, and not actually work "intrisically better".
69
 
70
  # Versions