Domain Adaptation from Synthesis to Reality in Single-model Detector for Video Smoke Detection
This work addresses video smoke detection for safety applications, but it is incremental as it builds on existing domain adaptation and detection methods.
The paper tackles the problem of video smoke detection by using synthetic smoke samples to train detectors, but addresses the appearance gap between synthetic and real smoke. The proposed method integrates domain adaptation directly into the detection layer of single-model detectors, achieving performance that surpasses the original baseline and showing adversarial adaptation outperforms discrepancy adaptation.
This paper proposes a method for video smoke detection using synthetic smoke samples. The virtual data can automatically offer precise and rich annotated samples. However, the learning of smoke representations will be hurt by the appearance gap between real and synthetic smoke samples. The existed researches mainly work on the adaptation to samples extracted from original annotated samples. These methods take the object detection and domain adaptation as two independent parts. To train a strong detector with rich synthetic samples, we construct the adaptation to the detection layer of state-of-the-art single-model detectors (SSD and MS-CNN). The training procedure is an end-to-end stage. The classification, location and adaptation are combined in the learning. The performance of the proposed model surpasses the original baseline in our experiments. Meanwhile, our results show that the detectors based on the adversarial adaptation are superior to the detectors based on the discrepancy adaptation. Code will be made publicly available on http://smoke.ustc.edu.cn. Moreover, the domain adaptation for two-stage detector is described in Appendix A.