Cvpr 2026 Decoupled Generative Modeling

[CVPR 2026] Decoupled Generative Modeling for Human-Object Interaction Synthesis

Presentation video for "

Even when you tell a diffusion model to "do nothing", it still changes your image. We call this No-Op Drift, and we prove it's not a ...

[

Text-guided diffusion models have advanced image editing by enabling intuitive control through language. However, despite their ...

Reinforcement Learning (RL) has achieved remarkable success in various domains, yet it often relies on carefully designed ...

Disentangle-then-Align: Non-Iterative Hybrid Multimodal Image Registration via Cross-Scale Feature Disentanglement.

In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,

Visual

CVPR 2026

[

We propose a federated framework for training diffusion models on decentralized and private datasets. The method learns a ...

Title:MU-GeNeRF: Multi-view Uncertainty-guided Generalizable Neural Radiance Fields for Distractor-aware Scene ...

Capturing geometry and rigid motion for structured dynamic objects is challenging. Existing methods rely on unstructured ...

[

Paper: https://arxiv.org/abs/2512.01390 Project Page: https://cmlab-korea.github.io/FRAMER/ Authors/Affiliations: [Seungho ...

DiffusionFF: A Diffusion-based Framework for Joint Face Forgery Detection and Fine-Grained Artifact Localization (

CVPR 2026

Chengxing Lin, Jinhong Deng, Yinjie Lei, Wen Li. "Deformation-based In-Context Learning for Point Cloud Understanding.