Mingchen Zhong

Poly-OPD: Heterogeneous Multi-Teacher On-Policy Distillation for Capability-Selectable Flow Models

Siming Fu, Haojun Xu, Ruizhe He, Zheming Fu et autres

Leading open text-to-image models often carry complementary strengths: one may lead on preference-aligned aesthetics while another follows compositional instructions more faithfully. However, differences in their autoencoders and noise schedules make it difficult to transfer these strengths across models. In this paper, we …

Poly-OPD: Heterogeneous Multi-Teacher On-Policy Distillation for Capability-Selectable Flow Models

Siming Fu, Haojun Xu, Ruizhe He, Zheming Fu et autres

Leading open text-to-image models often carry complementary strengths: one may lead on preference-aligned aesthetics while another follows compositional instructions more faithfully. However, differences in their autoencoders and noise schedules make it difficult to transfer these strengths across models. In this paper, we …

Égypte, cn (code pays fourni par la source)

Any-OPD: Heterogeneous On-Policy Distillation for Flow-Matching Models via Representation-Space Bridging

Siming Fu, Zheming Fu, Ruizhe He, Hualiang Wang et autres

On-policy distillation, in which a teacher corrects samples that the student itself generates, presupposes that the two models speak the same language: identical VAE latents, matching architectures, and a common timestep grid. We ask what happens when none of this holds, as …

OmniVR: Joint Video-Audio Conditional Generation for Restoring Degraded Historical Films

Xin Lu, Zihao Fan, Mingchen Zhong, Jie Huang et autres

Historical films suffer from co-occurring visual and audio degradations---blur, noise, flicker, hiss, clipping, and dropout---yet existing methods restore each modality independently, leaving quality gaps and cross-modal inconsistency. We present OmniVR, the first joint audio-video generative restoration model. Built upon a 22B-parameter audio-video …

Any-OPD: Heterogeneous On-Policy Distillation for Flow-Matching Models via Representation-Space Bridging

Siming Fu, Zheming Fu, Ruizhe He, Hualiang Wang et autres

On-policy distillation, in which a teacher corrects samples that the student itself generates, presupposes that the two models speak the same language: identical VAE latents, matching architectures, and a common timestep grid. We ask what happens when none of this holds, as …

OmniVR: Joint Video-Audio Conditional Generation for Restoring Degraded Historical Films

Xin Lu, Zihao Fan, Mingchen Zhong, Jie Huang et autres

Historical films suffer from co-occurring visual and audio degradations---blur, noise, flicker, hiss, clipping, and dropout---yet existing methods restore each modality independently, leaving quality gaps and cross-modal inconsistency. We present OmniVR, the first joint audio-video generative restoration model. Built upon a 22B-parameter audio-video …

Ultra Flash: Scaling Real-Time Streaming Video Generation to High Resolutions

Luxury, Jie Huang, Zihao Fan, Xiaoxiao Ma et autres

While recent autoregressive video diffusion models achieve remarkable streaming quality, they remain confined to low resolutions (e.g., 480P), leaving efficient, scalable, real-time high-resolution video generation a fundamental open challenge. To bridge this gap, we present Ultra Flash, a cascaded streaming framework capable …

Ultra Flash: Scaling Real-Time Streaming Video Generation to High Resolutions

Luxury, Jie Huang, Zihao Fan, Xiaoxiao Ma et autres

While recent autoregressive video diffusion models achieve remarkable streaming quality, they remain confined to low resolutions (e.g., 480P), leaving efficient, scalable, real-time high-resolution video generation a fundamental open challenge. To bridge this gap, we present Ultra Flash, a cascaded streaming framework capable …

CompEvent: Complex-valued Event-RGB Fusion for Low-light Video Enhancement and Deblurring

Mingchen Zhong, Xin Lu, Dong Li, Senyan Xu et autres

Low-light video deblurring poses significant challenges in applications like nighttime surveillance and autonomous driving due to dim lighting and long exposures. While event cameras offer potential solutions with superior low-light sensitivity and high temporal resolution, existing fusion methods typically employ staged strategies, …

cn (code pays fourni par la source)

Event-Conditioned Dual-Modal Fusion for Motion Deblurring

K. Liu, Mingchen Zhong, Senyan Xu, Zhijing Sun et autres

Motion deblurring aims to recover sharp frames from blurred inputs caused by camera shake or object movement during exposure. While deep learning-based methods have shown promising results, they often struggle with complex motion due to the absence of temporal cues in a …

cn (code pays fourni par la source)

NTIRE 2025 Challenge on Event-Based Image Deblurring: Methods and Results

Lei Sun, Andrea Alfarano, Shaolin Su, Kaiwei Wang et autres

This paper presents an overview of NTIRE 2025, the First Challenge on Event-Based Image Deblurring, detailing the proposed methodologies and corresponding results. The primary goal of the challenge is to design an event-based method that achieves high-quality image deblurring, with performance quantitatively …

NTIRE 2025 Challenge on Event-Based Image Deblurring: Methods and Results

Lei Sun, Andrea Alfarano, Shaolin Su, Kaiwei Wang et autres

This paper presents an overview of NTIRE 2025 the First Challenge on Event-Based Image Deblurring, detailing the proposed methodologies and corresponding results. The primary goal of the challenge is to design an event-based method that achieves high-quality image deblurring, with performance quantitatively …