Adaptive Color Grading paper: KNN beats end-to-end models at tonescale prediction

ssh4net · x · 2026-09-21

A paper by Trevor D. Canham, Abhijith Punnappurath, and Michael S. Brown introduces an open-source color grading tool and uses it to annotate a large dataset of video frames with tonescale region thresholds. Modeling experiments based on both practitioners' conventional wisdom and ML show K-nearest neighbors is an effective prediction strategy, outperforming state-of-the-art end-to-end approaches. Code is open source.

Original post →

More from Multimodal

Multimodal channel →