zahidzqj / flashocc

FlashOCC

FlashOcc: Fast and Memory-Efficient Occupancy Prediction via Channel-to-Height Plugin

News

• 2024.01.20 TensorRT Implement Writen In C++ With Cuda Acceleration
• 2023.12.23 Release the quick testing code via TensorRT in MMDeploy.
• 2023.11.28 Release the training code for FlashOcc.

This repository is an official implementation of FlashOCC

Given the capability of mitigating the long-tail deficiencies and intricate-shaped absence prevalent in 3D object detection, occupancy prediction has become a pivotal component in autonomous driving systems. However, the procession of three-dimensional voxel-level representations inevitably introduces large overhead in both memory and computation, obstructing the deployment of to-date occupancy prediction approaches. In contrast to the trend of making the model larger and more complicated, we argue that a desirable framework should be deployment-friendly to diverse chips while maintaining high precision. To this end, we propose a plug-and-play paradigm, namely FlashOCC, to consolidate rapid and memory-efficient occupancy prediction while maintaining high precision. Particularly, our FlashOCC makes two improvements based on the contemporary voxel-level occupancy prediction approaches. Firstly, the features are kept in the BEV, enabling the employment of efficient 2D convolutional layers for feature extraction. Secondly, a channel-to-height transformation is introduced to lift the output logits from the BEV into the 3D space. We apply the FlashOCC to diverse occupancy prediction baselines on the challenging Occ3D-nuScenes benchmarks and conduct extensive experiments to validate the effectiveness. The results substantiate the superiority of our plug-and-play paradigm over previous state-of-the-art methods in terms of precision, runtime efficiency, and memory costs, demonstrating its potential for deployment.

Main Results

Nuscenes Occupancy

1. FlashOcc on BEVDetOCC

Config	mIOU	FPS(Hz)	Flops(G)	Params(M)	Model	Log
BEVDetOCC-R50	31.60	92.1	241.76	29.02	gdrive	log
M0:FO(BEVDetOCC)-R50	31.95	197.6	154.1	39.94	gdrive	log
M1:FO(BEVDetOCC)-R50	32.08	152.7	248.57	44.74	gdrive	log
BEVDetOCC-R50-4D-Stereo	36.1	-	-	-	baidu	log
M2:FO(BEVDetOCC)-R50-4D-Stereo	37.84	-	-	-	gdrive	log
BEVDetOCC-STBase-4D-Stereo-512x1408	42.0	-	-	-	baidu	log
M3:FO(BEVDetOCC)-STBase-4D-Stereo-512x1408	43.52	-	1490.77	144.99	gdrive	log

2. FlashOcc on UniOCC, will come soon

3. FlashOcc on FB-OCC, will come soon

FPS are tested via TensorRT on 3090 with FP16 precision. Please refer to Tab.2 in paper for the detail model settings for M-number.

Get Started

1. Environment Setup
2. Model Training
3. Quick Test Via TensorRT In MMDeploy
4. Visualization
5. TensorRT Implement Writen In C++ With Cuda Acceleration

Acknowledgement

Many thanks to the authors of BEVDet, and the main code is based on it.

Bibtex

If this work is helpful for your research, please consider citing the following BibTeX entry.

@article{yu2023flashocc,
      title={FlashOcc: Fast and Memory-Efficient Occupancy Prediction via Channel-to-Height Plugin}, 
      author={Zichen Yu and Changyong Shu and Jiajun Deng and Kangjie Lu and Zongdai Liu and Jiangyong Yu and Dawei Yang and Hui Li and Yan Chen},
      year={2023},
      eprint={2311.12058},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}