arXivDaily arXiv每日学术速递 周一至周五更新
​通知:联邦节假日调休:哥伦布日 / 原住民日(Columbus Day / Indigenous Peoples' Day),今日(13日)无更新。
arXiv 2502.16652cs.CV

Dr. Splat:通过直接语言嵌入注册实现三维高斯溅射的直接引用

Dr. Splat: Directly Referring 3D Gaussian Splatting via Direct Language Embedding Registration

POSTECH(浦项科技大学);NVIDIA(英伟达);KAIST(韩国科学技术院)

Kim Jun-Seong, GeonU Kim, Kim Yu-Ji, Yu-Chiang Frank Wang, Jaesung Choe, Tae-Hyun Oh

AI总结:

Dr. Splat 提出一种直接注册语言嵌入到三维高斯的新方法,用于开放词汇三维场景理解,在分割、定位和选择任务上显著优于现有方法。

摘要:

我们提出了 Dr. Splat,一种利用三维高斯溅射进行开放词汇三维场景理解的新方法。与现有的依赖渲染过程的语言嵌入三维高斯溅射方法不同,我们的方法直接将语言对齐的 CLIP 嵌入与三维高斯关联,以实现整体的三维场景理解。我们方法的关键是一种语言特征注册技术,其中 CLIP 嵌入被分配给每条像素射线所穿过的显著高斯。此外,我们集成了在通用大规模图像数据上训练的乘积量化(PQ)来紧凑地表示嵌入,而无需逐场景优化。实验表明,我们的方法在三维感知基准测试中显著优于现有方法,例如开放词汇三维语义分割、三维对象定位和三维对象选择任务。视频结果请访问:https://drsplat.github.io/

英文摘要:

We introduce Dr. Splat, a novel approach for open-vocabulary 3D scene understanding leveraging 3D Gaussian Splatting. Unlike existing language-embedded 3DGS methods, which rely on a rendering process, our method directly associates language-aligned CLIP embeddings with 3D Gaussians for holistic 3D scene understanding. The key of our method is a language feature registration technique where CLIP embeddings are assigned to the dominant Gaussians intersected by each pixel-ray. Moreover, we integrate Product Quantization (PQ) trained on general large-scale image data to compactly represent embeddings without per-scene optimization. Experiments demonstrate that our approach significantly outperforms existing approaches in 3D perception benchmarks, such as open-vocabulary 3D semantic segmentation, 3D object localization, and 3D object selection tasks. For video results, please visit : https://drsplat.github.io/

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑