Zeqing Yuan

Zhekai Chen, Wen Wang, Zhen Yang et al.

We offer a novel approach to image composition, which integrates multiple input images into a single, coherent image. Rather than concentrating on specific use cases such as appearance editing (image harmonization) or semantic editing (semantic image composition), we showcase the potential of utilizing the powerful generative prior inherent in large-scale pre-trained diffusion models to accomplish generic image composition applicable to both scenarios. We observe that the pre-trained diffusion models automatically identify simple copy-paste boundary areas as low-density regions during denoising. Building on this insight, we propose to optimize the composed image towards high-density regions guided by the diffusion prior. In addition, we introduce a novel maskguided loss to further enable flexible semantic image composition. Extensive experiments validate the superiority of our approach in achieving generic zero-shot image composition. Additionally, our approach shows promising potential in various tasks, such as object removal and multiconcept customization.

Zeqing Leo Yuan, Mani Ramanagopal, Aswin C. Sankaranarayanan et al.

Decomposing an image into its intrinsic photometric factors--shading and reflectance--is a long-standing challenge due to the lack of extensive ground-truth data for real-world scenes. Recent methods rely on synthetic data or sparse annotations for limited indoor and even fewer outdoor scenes. We introduce a novel training-free approach for intrinsic image decomposition using only a pair of visible and thermal images. We leverage the principle that light not reflected from an opaque surface is absorbed and detected as heat by a thermal camera. This allows us to relate the ordinalities between visible and thermal image intensities to the ordinalities of shading and reflectance, which can densely self-supervise an optimizing neural network to recover shading and reflectance. We perform quantitative evaluations with known reflectance and shading under natural and artificial lighting, and qualitative experiments across diverse outdoor scenes. The results demonstrate superior performance over recent learning-based models and point toward a scalable path to curating real-world ordinal supervision, previously infeasible via manual labeling.