GIFT: Bootstrapping Image-to-CAD Program Synthesis via Geometric Feedback

TL;DR

GIFT is a novel data augmentation framework that enhances image-to-CAD program synthesis by leveraging geometric feedback, significantly improving robustness and efficiency without extra annotation or complex models.

Contribution

GIFT introduces a bootstrapping approach using geometric feedback, combining soft-rejection sampling and failure-driven augmentation to improve training data diversity and model robustness.

Findings

GIFT improves mean IoU by 12% over baseline.

Reduces inference compute by 80%.

Remains competitive without extra annotations or architectures.

Abstract

Generating executable CAD programs from images requires alignment between visual geometry and symbolic program representations, a capability that current methods fail to learn reliably as design complexity increases. Existing fine-tuning approaches rely on either limited supervised datasets or expensive post-training pipelines, resulting in brittle systems that restrict progress in generative CAD design. We argue that the primary bottleneck lies not in model or algorithmic capacity, but in the scarcity of diverse training examples that align visual geometry with program syntax. This limitation is especially acute because the collection of diverse and verified engineering datasets is both expensive and difficult to scale, constraining the development of robust generative CAD models. We introduce Geometric Inference Feedback Tuning (GIFT), a data augmentation framework that leverages geometric feedback to turn test-time compute into a bootstrapped set of high-quality training samples. GIFT combines two mechanisms: Soft-Rejection Sampling (GIFT-REJECT), which retains diverse high-fidelity programs beyond exact ground-truth matches, and Failure-Driven Augmentation (GIFT-FAIL), which converts near-miss predictions into synthetic training examples that improve robustness on challenging geometries. By amortizing inference-time search into the model parameters, GIFT captures the benefits of test-time scaling while reducing inference compute by 80%. It improves mean IoU by 12% over a strong supervised baseline and remains competitive with more complex multimodal systems, without requiring additional human annotation or specialized architectures.