Training-Time-Friendly Network for Real-Time Object Detection

By Zili Liu, Tu Zheng, Guodong Xu, Zheng Yang, Haifeng Liu, Deng Cai. 2019

98 citations recorded by the source corpus.

Modern object detectors can rarely achieve short training time, fast inference speed, and high accuracy at the same time. To strike a balance among them, we propose the Training-Time-Friendly Network (TTFNet). In this work, we start with light-head, single-stage, and anchor-free designs, which enable fast inference speed. Then, we focus on shortening training time. We notice that encoding more training samples from annotated boxes plays a similar role as increasing batch size, which helps enlarge the learning rate and accelerate the training process. To this end, we introduce a novel approach using Gaussian kernels to encode training samples. Besides, we design the initiative sample weights for better information utilization. Experiments on MS COCO show that our TTFNet has great advantages in balancing training time, inference speed, and accuracy. It has reduced training time by more than seven times compared to previous real-time detectors while maintaining state-of-the-art performances. In addition, our super-fast version of TTFNet-18 and TTFNet-53 can outperform SSD300 and YOLOv3 by less than one-tenth of their training time, respectively. The code has been made available at https://github.com/ZJULearning/ttfnet.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Hongpeng Liao
2. Chien-Yao Wang
3. I-Hau Yeh
4. Xinggang Wang
5. Wenyu Liu
6. Qian Zhang
7. Jiemin Fang
8. Yuan Li

Citations in the corpus

• Fast Neural Network Adaptation via Parameter Remapping and Architecture Search (2020, methodology): “For fair comparison with Auto-DeepLab (Liu et al., 2019a) which searches the back-bone architecture on DeepLabv3 and retrains the searched network on DeepLabv3+ (Chen et al., 2018b), we adapt the parameters of the target architecture Arch t to DeepLabv3+ framework.”
• CSPNet: A New Backbone that can Enhance Learning Capability of CNN (2019, methodology): “4 All results are obtained by COCO test-dev set except for TTFNet [22] models which are verified on minval5k set.”