6 papers
COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflows
Zongxia Li, Dawei Liu, Fuxiao Liu +6
Agents are increasingly used to construct workflows and assist humans in completing recurring tasks more efficiently. As these workflows become repeated and domain-specific, agent…
MASS: Motion-Aware Spatial-Temporal Grounding for Physics Reasoning and Comprehension in Vision-Language Models
Xiyang Wu, Zongxia Li, Jihui Jin +7
Vision Language Models (VLMs) perform well on standard video tasks but struggle with physics-related reasoning involving motion dynamics and spatial interactions. We present a nove…
From Inpainting to Layer Decomposition: Repurposing Generative Inpainting Models for Image Layer Decomposition
Jingxi Chen, Yixiao Zhang, Xiaoye Qian +4
Images can be viewed as layered compositions, foreground objects over background, with potential occlusions. This layered representation enables independent editing of elements, of…
First Frame Is the Place to Go for Video Content Customization
Jingxi Chen, Zongxia Li, Zhichao Liu +6
What role does the first frame play in video generation models? Traditionally, it's viewed as the spatial-temporal starting point of a video, merely a seed for subsequent animation…
Repurposing Pre-trained Video Diffusion Models for Event-based Video Interpolation
Jingxi Chen, Brandon Y. Feng, Haoming Cai +6
Video Frame Interpolation aims to recover realistic missing frames between observed frames, generating a high-frame-rate video from a low-frame-rate video. However, without additio…
Learning Normal Flow Directly From Event Neighborhoods
Dehao Yuan, Levi Burner, Jiayi Wu +4
Event-based motion field estimation is an important task. However, current optical flow methods face challenges: learning-based approaches, often frame-based and relying on CNNs, l…