← Back to all news
arXiv Computer Vision By Yujia Zhai (Great Bay University), Guoming Huang (Great Bay University), Zhuanghuai Wang (Great Bay University), Bangquan Xie (Great Bay University)

FM-Occ: Feature-Diverse Upsampling and Memory-Aware Decoding for 3D Occupancy Prediction

Read the original on arXiv Computer Vision →

FM-Occ introduces two novel components for 3D occupancy prediction: 3D Sparse Subvoxel Shuffle (3S) to diversify features during upsampling, and Entropy-Regulated Layer-wise Memory (ER‑LM) to preserve semantic information across decoder layers. 3S reduces feature homogenization by creating and rearranging child features on a finer sparse grid, while ER‑LM uses gated recurrent updates and entropy‑based regulation to maintain a memory of semantic predictions. On SemanticKITTI, FM‑Occ improves mIoU from 12.16% to 12.73% with minimal memory cost, and further boosts BEV‑based DISC performance from 16.78% to 17.17%.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

Related stories

arXiv Computer Vision
Aug 26

O3N: Omnidirectional Open-Vocabulary Occupancy Prediction for Embodied Intelligent Robotics

O3N is a novel framework that performs open‑vocabulary occupancy prediction from a single omnidirectional RGB image. It introduces a polar‑spiral voxel embedding (PsM) for continuous 360° spatial representation, an Occupancy Cost Aggregation (OCA) module that unifies geometric and semantic supervision, and a Natural Modality Alignment (NMA) pathway that aligns visual, voxel, and text features. Experiments show state‑of‑the‑art results on QuadOcc and Human360Occ benchmarks, with strong cross‑scene generalization and semantic scalability.

By Mengfei Duan, Hao Shi, Fei Teng, Guoqiang Zhao, Yuheng Zhang, Zhiyong Li, Kailun Yang
ragroboticsbenchmarkssafety
More like this →
arXiv AI
Oct 12

GeoGR: Spatial-Semantic Token Compression and Group Routing for Efficient 3D VLMs

GeoGR introduces a two‑stage framework for efficient 3D vision‑language models. The first stage, GeoSemZip, compresses multi‑view visual tokens into compact 3D voxel representations while preserving important scene information through contextual anchors. The second stage, GroupRoute, dynamically routes tokens through the language model, allowing low‑relevance tokens to skip layers and rejoin later, thereby reducing computation without significant loss of performance.

By Yongxin Zheng, Hao Wu, Fang Li, Hu Fu, Peiran Yin, Xudong Wang, Haozhe Hu, Xinghao Chen, Yunpu Ma, Wei Zhang, Xiaoyu Shen
llmsnlpmultimodal
More like this →