11 papers
HPSD: Hybrid-Policy Self-Distillation for Text-Image-to-Video Diffusion Models
Jiazi Bu, Pengyang Ling, Yujie Zhou +10
Text-Image-to-Video (TI2V) models are an emerging unified architecture, where a single model simultaneously supports text-to-video (T2V) and image-to-video (I2V) generation. Given…
The Surprising Effectiveness of Video Diffusion Models for Hand Motion Reconstruction
Yuxi Wang, Chengkai Jin, Yufei Liu +6
4D hand motion reconstruction from egocentric video is bottlenecked by clear limitations of existing methods: image-based pipelines depend on a detector that fails under heavy occl…
AdaGRPO: A Capability-Aware Adaptive Enhancement for Flow-based GRPO
Jiazi Bu, Pengyang Ling, Yujie Zhou +8
Group Relative Policy Optimization (GRPO) has demonstrated remarkable success in aligning text-to-image (T2I) flow models with human preferences. However, we have identified that t…
StructDiff: A Structure-Preserving and Spatially Controllable Diffusion Model for Single-Image Generation
Yinxi He, Kang Liao, Chunyu Lin +2
This paper introduces StructDiff, a generative framework based on a single-scale diffusion model for single-image generation. Single-image generation aims to synthesize diverse sam…
From Sparse to Dense: Multi-View GRPO for Flow Models via Augmented Condition Space
Jiazi Bu, Pengyang Ling, Yujie Zhou +8
Group Relative Policy Optimization (GRPO) has emerged as a powerful framework for preference alignment in text-to-image (T2I) flow models. However, we observe that the standard par…
Hand2World: Autoregressive Egocentric Interaction Generation via Free-Space Hand Gestures
Yuxi Wang, Wenqi Ouyang, Tianyi Wei +3
Egocentric interactive world models are essential for augmented reality and embodied AI, where visual generation must respond to user input with low latency, geometric consistency,…
PnP-U3D: Plug-and-Play 3D Framework Bridging Autoregression and Diffusion for Unified Understanding and Generation
Yongwei Chen, Tianyi Wei, Yushi Lan +4
The rapid progress of large multimodal models has inspired efforts toward unified frameworks that couple understanding and generation. While such paradigms have shown remarkable su…
PI-Light: Physics-Inspired Diffusion for Full-Image Relighting
Zhexin Liang, Zhaoxi Chen, Yongwei Chen +3
Full-image relighting remains a challenging problem due to the difficulty of collecting large-scale structured paired data, the difficulty of maintaining physical plausibility, and…
Trainable Log-linear Sparse Attention for Efficient Diffusion Transformers
Yifan Zhou, Zeqi Xiao, Tianyi Wei +2
Diffusion Transformers (DiTs) set the state of the art in visual generation, yet their quadratic self-attention cost fundamentally limits scaling to long token sequences. Recent To…
Boosting Monocular Metric Depth Estimation via Bokeh Rendering
Hangwei Zhang, Armando Fortes, Tianyi Wei +1
Bokeh rendering and depth estimation share a fundamental optical connection, yet existing methods fail to fully exploit this reciprocity. Conventional bokeh pipelines rely heavily…
Towards Geometry-Grounded Dense Semantic Matching with VGGT Priors
Songlin Yang, Tianyi Wei, Yushi Lan +3
Semantic matching aims to establish pixel-level correspondences between instances of the same category and represents a fundamental task in computer vision. Existing approaches suf…