“Physically-Based Compositing of 2D Graphics” by Tracey, Diaconu, Dille, Miangoleh and Aksoy
Conference:
Type(s):
Title:
- Physically-Based Compositing of 2D Graphics
Session/Category Title:
- Images, Video & Computer Vision
Presenter(s)/Author(s):
Abstract:
We propose a geometry- and illumination-aware 2d-graphic compositing pipeline. We use meshes generated by off-the-shelf monocular depth estimation methods to warp the 2d-graphic according to the surface geometry. Using intrinsic decomposition, we composite the warped graphic onto the albedo and reconstruct the final result by combining all intrinsic components.
References:
[1] Blender. 2025. Blender – a 3D modelling and rendering package. http://www.blender.org
[2] Chris Careaga and Yağız Aksoy. 2024. Colorful Diffuse Intrinsic Image Decomposition in the Wild. ACM Trans. Graph. 43, 6, Article 178 (2024), 12 pages.
[3] Kalash Jain. 2024. utils3d. https://pypi.org/project/utils3d/
[4] Sophie J. Nightingale, Kimberly A. Wade, and Derrick G. Watson. 2022. Investigating age-related differences in ability to distinguish between original and manipulated images. Psychology and Aging 37, 3 (2022), 11 pages.
[5] Ruicheng Wang, Sicheng Xu, Cassie Dai, Jianfeng Xiang, Yu Deng, Xin Tong, and Jiaolong Yang. 2025. MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervision. In Proc. CVPR.
