Submitted by Haomeng Zhang.

Submission data

Full nameD-LISA
Publication titleMulti-Object 3D Grounding with Dynamic Modules and Language-Informed Spatial Attention
Publication authorsHaomeng Zhang, Chiao-An Yang, Raymond A. Yeh
Publication venueNeurIPS 2024
Publication URLhttps://arxiv.org/abs/2410.22306
Input Data TypesUses XYZ coordinates,Uses Multiview Image Features
Programming language(s)Pytorch
HardwareA100
Websitehttps://haomengz.github.io/dlisa
Submission creation date2 May, 2024
Last edited1 Nov, 2024

Localization

Unique Unique Multiple Multiple Overall Overall
acc@0.25IoUacc@0.5IoUacc@0.25IoUacc@0.5IoUacc@0.25IoUacc@0.5IoU
0.81950.69000.49750.39670.56970.4625