demo.git - Vide-coded 64k demo system

Age	Commit message (Expand)	Author
28 hours	feat: Add salient-point patch extraction for CNN training	skal
29 hours	fix: Correct UV coordinate computation to match PyTorch linspace	skal
29 hours	fix: Add clamp to CNN final layer to match PyTorch training	skal
30 hours	refactor: Optimize CNN grayscale computation	skal
30 hours	update train_cnn.py and shader	skal
31 hours	feat: Add inference mode to train_cnn.py for ground truth generation	skal
31 hours	fix: CNN training normalization pipeline consistency	skal
32 hours	refactor: Optimize CNN normalization to eliminate redundant conversions	skal
33 hours	fix: Support variable kernel sizes in CNN layer generation	skal
34 hours	feat: CNN RGBD→grayscale with 7-channel augmented input	skal
38 hours	feat: Add multi-layer CNN support with framebuffer capture and blend control	skal
40 hours	feat: Add checkpointing support to CNN training script	skal
40 hours	fix: Auto-expand single kernel size to all layers in training script	skal
40 hours	feat: Add coordinate-aware CNN layer 0 for position-dependent stylization	skal