CNN v2: parametric static features - design doc

Design document for CNN v2 with enhanced feature inputs:
- 7D static features: RGBD + UV + sin encoding + bias
- Per-layer configurable kernels (1×1, 3×3, 5×5)
- Float16 weight storage (~6.4 KB vs 3.2 KB)
- Multi-pass architecture with static feature compute

Implementation plan:
1. Static features compute shader (RGBD + UV + sin + bias)
2. C++ effect class (CNNv2Effect)
3. Training pipeline (train_cnn_v2.py, export_cnn_v2_shader.py)
4. Validation tooling (validate_cnn_v2.sh)

Files:
- doc/CNN_V2.md: Complete technical design (architecture, training, export)
- scripts/validate_cnn_v2.sh: End-to-end validation script
- TODO.md: Add CNN v2 as Priority 2 task
- doc/HOWTO.md: Add CNN v2 validation usage

Target: <10 KB for 64k demo constraint

handoff(Claude): CNN v2 design ready for implementation

1 files changed, 21 insertions, 0 deletions

diff --git a/TODO.md b/TODO.md
index d7d24bc..b0cf2bb 100644
--- a/TODO.md
+++ b/TODO.md
@@ -24,6 +24,27 @@ Self-contained workspaces for parallel demo development.
 
 ---
 
+## Priority 2: CNN v2 - Parametric Static Features (Task #85) [PLANNING]
+
+Enhanced CNN post-processing with multi-dimensional feature inputs.
+
+**Design:** `doc/CNN_V2.md`
+
+**Implementation phases:**
+1. Static features compute shader (RGBD + UV + sin encoding + bias)
+2. C++ effect class (multi-pass layer execution)
+3. Training pipeline (PyTorch f32 → f16 export)
+4. Validation tooling (end-to-end checkpoint testing)
+
+**Key improvements over v1:**
+- 7D static feature input (vs 4D RGB)
+- Per-layer configurable kernels (1×1, 3×3, 5×5)
+- Float16 weight storage (~6.4 KB vs 3.2 KB)
+
+**Target:** <10 KB for 64k demo constraint
+
+---
+
 ## Priority 3: 3D System Enhancements (Task #18)
 
 Pipeline for importing complex 3D scenes to replace hardcoded geometry.