Slim-neck by GSConv: a lightweight-design for real-time detector architectures

• Published in: Journal of real-time image processing Vol. 21; no. 3; p. 62
• Main Authors: Li, Hulin, Li, Jun, Wei, Hanbing, Liu, Zheng, Zhan, Zhenfei, Ren, Qiliang
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.05.2024; Springer Nature B.V
• Subjects: Accuracy; Computational efficiency; Computer Graphics; Computer Science; Design; Detectors; Effectiveness; Image Processing and Computer Vision; Lightweight; Multimedia Information Systems; Neural networks; Object recognition; Pattern Recognition; Real time; Sensors; Signal,Image and Speech Processing; Weight reduction; Deep learning; GSConv; Real-time detection; Lightweight; CNNs; Edge computing
• ISSN: 1861-8200; 1861-8219
• DOI: 10.1007/s11554-024-01436-6

Real-time object detection is significant for industrial and research fields. On edge devices, a giant model is difficult to achieve the real-time detecting requirement, and a lightweight model built from a large number of the depth-wise separable convolutional could not achieve the sufficient accuracy. We introduce a new lightweight convolutional technique, GSConv, to lighten the model but maintain the accuracy. The GSConv accomplishes an excellent trade-off between the accuracy and speed. Furthermore, we provide a design suggestion based on the GSConv, slim-neck (SNs), to achieve a higher computational cost-effectiveness of the real-time detectors. The effectiveness of the SNs was robustly demonstrated in over twenty sets comparative experiments. In particular, the real-time detectors of ameliorated by the SNs obtain the state-of-the-art (70.9% AP 50 for the SODA10M at a speed of ~ 100 FPS on a Tesla T4) compared with the baselines. Code is available at https://github.com/alanli1997/slim-neck-by-gsconv .