no code implementations • 22 Jan 2024 • Leichao Cui, Xiuxian Li, Min Meng, Guangyu Jia
To tackle the challenge of expanding the receptive field of a 3D convolutional kernel, we introduce the Dynamic Feature Fusion Module (DFFM).
no code implementations • 13 Mar 2023 • Leichao Cui, Xiuxian Li, Min Meng, Xiaoyu Mo
However, due to the sparsity of point clouds, accurate and reliable detection cannot be achieved in specific scenarios.