paper-with-me

Papers

Enhancing Scene Transition Awareness in Video Generation via Post-Training

2025-07-24 · Hanwen Shen, Jiajie Lu, Yupeng Cao, Xiaonan Yang arxiv

Recent advances in AI-generated video have shown strong performance on \emph{text-to-video} tasks, particularly for short clips depicting a single scene. However, current models struggle to generate longer videos with coherent scene transitions, primarily because they cannot infer when a transition is needed from the prompt. Most open-source models are trained on datasets consisting of single-scene video clips, which limits their capacity to learn and respond to prompts requiring multiple scenes. Developing scene transition awareness is essential for multi-scene generation, as it allows models to identify and segment videos into distinct clips by accurately detecting transitions. To address this, we propose the \textbf{Transition-Aware Video} (TAV) dataset, which consists of preprocessed video clips with multiple scene transitions. Our experiment shows that post-training on the \textbf{TAV} dataset improves prompt-based scene transition understanding, narrows the gap between required and generated scenes, and maintains image quality.

📄 PDF Abstract BibTeX arXiv:2507.18046

Code (0)

등록된 구현이 없습니다.

Tasks

Scene GenerationVideo Generation

Similar Papers 제목 키워드 기반

Trans4D: Realistic Geometry-Aware Transition for Compositional Text-to-4D Synthesis

2024-10-09 · Bohan Zeng, Ling Yang, Siyu Li, Jiaming Liu 외

Recent advances in diffusion models have demonstrated exceptional capabilities in image and video generation, further improving the effectiveness of 4D synthesis. Existing 4D generation methods can generate high-quality …

Video Generation

Versatile Transition Generation with Image-to-Video Diffusion

2025-08-03 · Zuhao Yang, Jiahui Zhang, Yingchen Yu, Shijian Lu 외 arxiv

Leveraging text, images, structure maps, or motion trajectories as conditional guidance, diffusion models have achieved great success in automated and high-quality video generation. However, generating smooth and rationa…

Video Generation

VSTAR: A Video-grounded Dialogue Dataset for Situated Semantic Understanding with Scene and Topic Transitions

2023-05-30 · Yuxuan Wang, Zilong Zheng, Xueliang Zhao, Jinpeng Li 외

Video-grounded dialogue understanding is a challenging problem that requires machine to perceive, parse and reason over situated semantics extracted from weakly aligned video and dialogues. Most existing benchmarks treat…

Dialogue GenerationDialogue UnderstandingScene SegmentationSegmentation

SEINE: Short-to-Long Video Diffusion Model for Generative Transition and Prediction

2023-10-31 · Xinyuan Chen, Yaohui Wang, Lingjun Zhang, Shaobin Zhuang 외

Recently video generation has achieved substantial progress with realistic results. Nevertheless, existing AI-generated videos are usually very short clips ("shot-level") depicting a single scene. To deliver a coherent l…

PredictionSemantic SimilaritySemantic Textual SimilarityVideo Generation+1

Do Joint Audio-Video Generation Models Understand Physics?

2026-05-08 · Zijun Cui, Xiulong Liu, Hao Fang, Mingwei Xu 외 arxiv

Joint audio-video generation models are rapidly approaching professional production quality, raising a central question: do they understand audio-visual physics, or merely generate plausible sounds and frames that violat…

Video Generation