Tao He, Jianbing Shen, Dubing Chen, Huan Zheng, Wenlong Liao, Pai Peng, Xianfei Li, Jin Fang, Xingping Dong
We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. If it links a repository, it is listed below.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
We present GDFusion, a temporal fusion method for vision-based 3D semantic occupancy prediction (Vi-sionOcc). GDFusion opens up the underexplored aspects of temporal fusion within the VisionOcc framework, focusing on both temporal cues and fusion strategies. It systematically examines the entire VisionOcc pipeline, identifying three fundamental yet previously overlooked temporal cues: scene-level consistency, motion calibration, and geometric complementation. These cues capture diverse facets of temporal evolution and make distinct contributions across various modules in the VisionOcc framework. To effectively fuse temporal signals across heterogeneous representations, we propose a novel fusion strategy by reinterpreting the formulation of vanilla RNNs. This reinterpretation leverages gradient descent on features to unify the integration of diverse temporal information, seamlessly embedding the proposed temporal cues into the network. Extensive experiments on nuScenes demonstrate that GDFusion significantly outperforms established baselines, achieving 2.2%-4.7% mIoU improvement and reducing memory consumption by 30%-72%. Codes are available at https: //github.com/cdb342/GDFusion.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2504.12959")
get_code_for_paper("2504.12959")
have("2504.12959")
Connect an agent — have() is free.