Artificial Intelligence For the Stereographer

Описание к видео Artificial Intelligence For the Stereographer

Originally presented at the NSA Virtual 3D-Con 2021 (http://3d-con.com), 8/15/21
The collected links: http://WorldOfDepth.com/tutorials/AIf...

Many modern developments in AI deal with the 3rd dimension, from self-driving cars that perceive and navigate their environments in 3D, to programs that can build 3D models of scenes simply by analyzing 2D videos of them. Though some of these AIs are only accessible to researchers and tech companies, some AI applications like upscaling images or colorizing black and white photos are immediately useful and available to all. This workshop will show you how to use AIs of special interest to stereographers, via the free tool Google Colab, to do things like creating depth maps, stereopairs, animations, and stereovideos—even from 2D images.

UPSCALING IMAGES
Recommended: http://BigJPG.com

Neural Image Super-Resolution by Christian Ledig et al. *not covered in this workshop: http://github.com/fukumame/superresol...
• http://colab.research.google.com/gith...

COLORIZING BLACK & WHITE IMAGES
Website version: http://deepai.org/machine-learning-mo...
DeOldify by Jason Antic et al.: http://github.com/jantic/DeOldify
• Artistic version (for more interesting details, and vibrance)
Customized for direct uploading: https://colab.research.google.com/dri...
Original: http://colab.research.google.com/gith...
• Stable version (better for landscapes and portraits)
Original: http://colab.research.google.com/gith...

CREATING DEPTH MAPS FROM 2D IMAGES
Boosting Monocular Depth Estimation Models to High-Resolution via Content-Adaptive Multi-Resolution Merging by S. Mahdi H. Miangoleh et al.:
http://yaksoy.github.io/highresdepth/
• Original: http://colab.research.google.com/gith...
• Bug fix version (if the above gives you an error): https://colab.research.google.com/dri...
Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer by René Ranftl et al.: http://github.com/isl-org/MiDaS
• Customized for direct uploading: https://colab.research.google.com/dri...
• Original: http://colab.research.google.com/gith...

USING DEPTH MAPS TO ANIMATE 2D IMAGES
3D Photography using Context-aware Layered Depth Inpainting by Meng-Li Shih et al.: http://shihmengli.github.io/3D-Photo-...
• https://colab.research.google.com/dri...
• Documentation/Settings: http://github.com/vt-vl-lab/3d-photo-...

EXTRAPOLATING DEPTH MAPS FROM 2D VIDEOS
Consistent Video Depth Estimation by Xuan Lo et al.: http://roxanneluo.github.io/Consisten...
•    • Consistent Video Depth Estimation  
• https://colab.research.google.com/dri... *NEW / not covered in this workshop
Neural Scene Flow Fields for Space-Time View Synthesis of Dynamic Scenes by Zhengqi Li et al.: http://www.cs.cornell.edu/~zl548/NSFF/
•    • Neural Scene Flow Fields for Space-Ti...  

The original stereophotos were captured with:
• my DIY catadioptric rig (see http://worldofdepth.com/daily/201023....)
• twin Sony RX100.ii rig
• 3D LG Thrill
• sequentially with an iPhone SE

The stereos were edited using:
• StereoPhoto Maker: http://stereo.jpg.org
• http://ImageMagick.org
• http://Photopea.com
• http://ffmpeg.org (for videos)

http://WorldOfDepth.com
  / worldofdepth  

Written/produced by Gordon Au © 2021

Комментарии

Информация по комментариям в разработке