Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces a method for precise camera control in text-to-image generation by learning parametric camera tokens through fine-tuning on a dataset combining 3D-rendered images and photorealistic augmentations. The learned viewpoint tokens capture factorized geometric representations that generalize to novel object categories, overcoming limitations of previous methods that overfit object-specific appearance correlations. Experiments demonstrate state-of-the-art accuracy in viewpoint control while maintaining image quality and prompt fidelity.
Text-to-image models can now learn explicit 3D camera control from language, unlocking geometrically-aware prompts that generalize beyond object-specific biases.
Current text-to-image models struggle to provide precise camera control using natural language alone. In this work, we present a framework for precise camera control with global scene understanding in text-to-image generation by learning parametric camera tokens. We fine-tune image generation models for viewpoint-conditioned text-to-image generation on a curated dataset that combines 3D-rendered images for geometric supervision and photorealistic augmentations for appearance and background diversity. Qualitative and quantitative experiments demonstrate that our method achieves state-of-the-art accuracy while preserving image quality and prompt fidelity. Unlike prior methods that overfit to object-specific appearance correlations, our viewpoint tokens learn factorized geometric representations that transfer to unseen object categories. Our work shows that text-vision latent spaces can be endowed with explicit 3D camera structure, offering a pathway toward geometrically-aware prompts for text-to-image generation. Project page: https://randdl.github.io/viewtoken_control/