SimDistill: Simulated Multi-Modal Distillation for BEV 3D Object Detection

Authors: Haimei Zhao, Qiming Zhang, Shanshan Zhao, Zhe Chen, Jing Zhang, Dacheng Tao

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments validate the effectiveness and superiority of Sim Distill over state-of-the-art methods, achieving an improvement of 4.8% m AP and 4.1% NDS over the baseline detector. The source code will be released at https://github.com/Vi TAE-Transformer/Sim Distill.
Researcher Affiliation Academia 1School of Computer Science, The University of Sydney, Australia, 2School of Computing, Engineering and Mathematical Sciences, La Trobe University, Australia
Pseudocode No The paper does not contain any pseudocode or algorithm blocks.
Open Source Code Yes The source code will be released at https://github.com/Vi TAE-Transformer/Sim Distill.
Open Datasets Yes We follow the common practice (Huang et al. 2021; Liu et al. 2023; Liang et al. 2022; Li et al. 2023b; Chen et al. 2023) to evaluate our method on the most challenging benchmark, i.e., nu Scenes (Caesar et al. 2020).
Dataset Splits Yes It comprises 700 scenes for training, 150 scenes for validation, and 150 scenes for testing.
Hardware Specification Yes Our method is implemented with Py Torch using 8 NVIDIA A100 (40G Memory), based on the MMDetection3D codebase (Contributors 2020).
Software Dependencies No The paper mentions "Py Torch" and "MMDetection3D codebase" but does not specify their version numbers.
Experiment Setup Yes train the student model for 20 epochs with batch size 24.