RT @ZhiqiuLin: 📷 Can AI understand camera motion like a cinematographer?. Meet CameraBench: a large-scale, expert-annotated dataset for und….

RT @ZhiqiuLin: Fresh GPT‑o3 results on our vision‑centric #NaturalBench (NeurIPS’24) benchmark! 🎯 Its new visual chain‑of‑thought—by “zoomi….

RT @kvuongdev: [1/6] Recent models like DUSt3R generalize well across viewpoints, but performance drops on aerial-ground pairs. At #CVPR20….

RT @chancharikm: 🎯 Introducing Sparse Attention Vectors (SAVs): A breakthrough method for extracting powerful multimodal features from Larg….

RT @ZhiqiuLin: 🚀 Make Vision Matter in Visual-Question-Answering (VQA)!. Introducing NaturalBench, a vision-centric VQA benchmark (NeurIPS'….

RT @anirudhchak: Lidar Panoptic Segmentation (LPS) is crucial for the safe deployment of autonomous vehicles, but it fails to consider real….

RT @kangle_deng: [1/2].📢 I'll present "FlashTex: Fast Relightable Mesh Texturing with LightControlNet" at #ECCV2024’s Oral Session tomorrow….

RT @ZhiqiuLin: Sharing exciting news from Milan 🇮🇹: VQAScore (ECCV’24) was highlighted as the strongest text-to-image metric in DeepMind’s….

RT @tarashakhurana: Our new work explores generating future observations (blue) given the past (gray), by leveraging large-scale pretrainin….

RT @ZhiqiuLin: In text-to-image generation, evaluating how well the generated image matches the prompt is a major challenge. We address thi….