3 papers
cs.CV2025
Q-CLIP: Unleashing the Power of Vision-Language Models for Video Quality Assessment through Unified Cross-Modal Adaptation
Yachun Mi, Yu Li, Yanting Li +6
Accurate and efficient Video Quality Assessment (VQA) has long been a key research challenge. Current mainstream VQA methods typically improve performance by pretraining on large-s…
cs.CV2025
Refining CLIP's Spatial Awareness: A Visual-Centric Perspective
Congpei Qiu, Yanhao Wu, Wei Ke +2
Contrastive Language-Image Pre-training (CLIP) excels in global alignment with language but exhibits limited sensitivity to spatial information, leading to strong performance in ze…
cs.GR2024
Scaling Mesh Generation via Compressive Tokenization
Haohan Weng, Zibo Zhao, Biwen Lei +10
We propose a compressive yet effective mesh representation, Blocked and Patchified Tokenization (BPT), facilitating the generation of meshes exceeding 8k faces. BPT compresses mesh…