Reference-based sketch colorization methods rely on large paired datasets that preserve both the structural and stylistic characteristics of hand-drawn artwork. However, existing datasets are limited in scale, expensive to annotate, and bound to fixed, often inconsistent artistic style biases that propagate to downstream models and limit cross-domain generalization. We present TexSketch, a controllable procedural framework for generating colored-sketch datasets with programmable artistic styles via geometric analysis and shader-driven stylization. Our fully automatic pipeline integrates region extraction, semantic color prediction, and shader-based rendering. By defining artistic appearance procedurally rather than inheriting it from a static corpus, TexSketch enables scalable dataset generation without manual annotation or artist supervision. Human studies demonstrate that TexSketch generates perceptually plausible colored sketches with high stylistic diversity, providing a controllable, scalable source of synthetic supervision for sketch colorization.
An overview of our method. Our method involves the following steps: (a) Segment an input sketch. (b) For each segment, figure out its likely hue by overlaying the segment on the image and querying a VLM. (c) Run the procedural framework to get geometric information and affect the color. (d) Finally, combine all textured segments and run post-processing.
Comparison of TexSketch with Colorize Diffusion XL on five input sketches. Drag the handle to compare. Note: ColorizeDiffusion always returns images in 512x512 while our method is resolution independent. The images you see below are truncated to fit within 512x512 pixels.
TexSketch preserves the input sketch while allowing the rendered appearance to be controlled with different styles. Drag either handle to compare all three styles side by side.
This research work is supported by the Center for Design and New Media (A TCS Foundation Initiative supported by Tata Consultancy Services) at IIIT-Delhi, India.
