PAR3D: A Unified 3D-MLLM with Part-Aware Representation for Scene Understanding
PAR3D: 一种用于场景理解的统一部件感知3D多模态大语言模型
机构 * Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University(教育部多媒体可信感知与高效计算重点实验室,厦门大学)
专题命中 视觉问答 :MLLM(title,summary_cn);visual question answering(abstract);multimodal large language model(abstract);分类 cs.CV
AI总结 提出PAR3D框架,通过部件感知3D表示学习和层次化分割查询生成,解决现有3D-MLLM在细粒度部件理解上的不足,在部件级问答和指代分割任务上取得显著提升。
Comments Project page: https://atrovast.github.io/PAR3D/