TediGAN
clip-glass
| TediGAN | clip-glass | |
|---|---|---|
| 1 | 13 | |
| 361 | 177 | |
| 0.0% | - | |
| 0.0 | 0.0 | |
| about 1 year ago | over 2 years ago | |
| Python | Python | |
| MIT License | GNU General Public License v3.0 only |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
TediGAN
-
test
(Added Feb. 23, 2021) TediGAN - Colaboratory by weihaox. Uses StyleGAN to generate images. GitHub. I got error "No pre-trained weights found for perceptual model!" when I used the Colab notebook, which was fixed when I made the change mentioned here. After this change, I still got an error in the cell that displays the images, but the results were in the remote file system. Use the "Files" icon on the left to browse the remote file system.
clip-glass
-
test
(Added Feb. 5, 2021) CLIP-GLaSS.ipynb - Colaboratory by Galatolo. Uses BigGAN (default) or StyleGAN to generate images. The GPT2 config is for image-to-text, not text-to-image. GitHub.
-
Image to text models
After a cursory search I found CLIP-GLaSS and CLIP-cap. I've used CLIP-GLaSS in a previous experiment, but found the captions for digital/CG images quite underwhelming. This is understandable since this is not what the model was trained on, but still I'd like to use a better model.
-
[R] end-to-end image captioning
CLIP-GLaSS
- What CLIP-GLaSS thinks Ancient Egyptian computers would look like
-
Texttoimage 3 Images For Text Photo Of Donald
The images were generated using this notebook.
- CLIP-GLaSS prompt: "Screenshot of a video game from the 1930s"
-
[P] List of sites/programs/projects that use OpenAI's CLIP neural network for steering image/video creation to match a text description
The CLIP-GLaSS project has image-to-text functionality (I haven't tried it.)
-
For educational purposes: Text-to-image (3 runs with no cherry-picking, 6 images each) for text "Photo of a Lamborghini painted purple and red" generated using CLIP-GLaSS. config=StyleGAN2_car_d. save_each=50. generations=1000
Link to notebook.
-
Sharing CLIP magic based on OpenAI's blog post via a bit more accessible YT medium. Lmk what u think 🙈 ❤️
CLIP-GLaSS
-
[R] [P] Generating images from caption and vice versa via CLIP-Guided Generative Latent Space Search. Link to code and Google Colab notebook for project CLIP-GLaSS is in a comment.
Github for CLIP-GLaSS is here.
What are some alternatives?
StyleCLIP - Using CLIP and StyleGAN to generate faces from prompts.
a-PyTorch-Tutorial-to-Image-Captioning - Show, Attend, and Tell | a PyTorch Tutorial to Image Captioning
Colab-deep-daze - Simple command line tool for text to image generation using OpenAI's CLIP and Siren (Implicit neural representation network)
meshed-memory-transformer - Meshed-Memory Transformer for Image Captioning. CVPR 2020
VectorAscent - Generate vector graphics from a textual caption
deep-daze - Simple command line tool for text to image generation using OpenAI's CLIP and Siren (Implicit neural representation network). Technique was originally created by https://twitter.com/advadnoun
AuViMi - AuViMi stands for audio-visual mirror. The idea is to have CLIP generate its interpretation of what your webcam sees, combined with the words thare are spoken.
aphantasia - CLIP + FFT/DWT/RGB = text to image/video
Story2Hallucination
stylized-neural-painting - Official Pytorch implementation of the preprint paper "Stylized Neural Painting", in CVPR 2021.
clipping-CLIP-to-GAN