notesum.ai

Published at December 5

EmbodiedOcc: Embodied 3D Occupancy Prediction for Vision-based Online Scene Understanding

cs.CV

cs.AI

cs.LG

Released Date: December 5, 2024

Authors: Yuqi Wu¹, Wenzhao Zheng¹, Sicheng Zuo¹, Yuanhui Huang¹, Jie Zhou¹, Jiwen Lu¹

Aff.: ¹Tsinghua University, China

Arxiv: http://arxiv.org/pdf/2412.04380v1

[Uncaptioned image]

Method	Dataset	IoU	$\blacksquare$ ceiling	$\blacksquare$ floor	$\blacksquare$ wall	$\blacksquare$ window	$\blacksquare$ chair	$\blacksquare$ bed	$\blacksquare$ sofa	$\blacksquare$ table	$\blacksquare$ tvs	$\blacksquare$ furniture	$\blacksquare$ objects	mIoU
SplicingOcc	EmbodiedOcc	49.01	31.60	38.80	35.50	36.30	47.10	54.50	57.20	34.40	32.50	51.20	29.10	40.74
EmbodiedOcc	EmbodiedOcc	51.52	22.70	44.60	37.40	38.00	50.10	56.70	59.70	35.40	38.40	52.00	32.90	42.53
SplicingOcc	EmbodiedOcc-mini	48.75	29.00	37.60	37.30	26.80	44.50	65.90	52.70	40.80	36.60	54.50	27.90	41.24
EmbodiedOcc	EmbodiedOcc-mini	50.78	22.10	43.70	39.00	26.60	45.00	63.70	54.40	43.90	34.70	55.30	27.60	41.45