Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
AceTone is introduced, a novel approach to conditional image color grading that generates 3D lookup tables (LUTs) based on text prompts or reference images. A VQ-VAE tokenizer is used to compress LUTs into discrete tokens, enabling a vision-language model to predict these tokens based on the conditioning input. Reinforcement learning is then applied to align the generated LUTs with perceptual fidelity and aesthetic preferences, resulting in state-of-the-art performance on both text-guided and reference-guided grading tasks.
AceTone lets you steer image color grading with language, achieving a 50% LPIPS improvement and opening new creative possibilities.
Color affects how we interpret image style and emotion. Previous color grading methods rely on patch-wise recoloring or fixed filter banks, struggling to generalize across creative intents or align with human aesthetic preferences. In this study, we propose AceTone, the first approach that supports multimodal conditioned color grading within a unified framework. AceTone formulates grading as a generative color transformation task, where a model directly produces 3D-LUTs conditioned on text prompts or reference images. We develop a VQ-VAE based tokenizer which compresses a $3\times32^3$ LUT vector to 64 discrete tokens with $\Delta E<2$ fidelity. We further build a large-scale dataset, AceTone-800K, and train a vision-language model to predict LUT tokens, followed by reinforcement learning to align outputs with perceptual fidelity and aesthetics. Experiments show that AceTone achieves state-of-the-art performance on both text-guided and reference-guided grading tasks, improving LPIPS by up to 50% over existing methods. Human evaluations confirm that AceTone's results are visually pleasing and stylistically coherent, demonstrating a new pathway toward language-driven, aesthetic-aligned color grading.