Choi, JunYong Sagong, Min-Cheol Lee, SeokYeong Jung, Seung-Won Kim, Ig-Jae Cho, Junghyun 2025-10-31T01:30:28Z 2025-10-31T01:30:28Z 2025-10-20 2025-06-10 https://pubs.kist.re.kr/handle/201004/153382 We propose a diffusion-based inverse rendering framework that decomposes a single RGB image into geometry, material, and lighting. Inverse rendering is inherently ill-posed, making it difficult to predict a single accurate solution. To address this challenge, recent generative model-based methods aim to present a range of possible solutions. However, finding a single accurate solution and generating diverse solutions can be conflicting. In this paper, we propose a channel-wise noise scheduling approach that allows a single diffusion model architecture to achieve two conflicting objectives. The resulting two diffusion models, trained with different channel-wise noise schedules, can predict a single highly accurate solution and present multiple possible solutions. The experimental results demonstrate the superiority of our two models in terms of both diversity and accuracy, which translates to enhanced performance in downstream applications such as object insertion and material editing. English IEEE COMPUTER SOC Channel-wise Noise Scheduled Diffusion for Inverse Rendering in Indoor Scenes Conference 10.1109/cvpr52734.2025.00542 1 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp.5773 - 5782 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 5773 5782 US Nashville, TN, USA 2025-06-10 2025 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 001562507806020