Harmonizing Transferability and Discriminability for Adapting Object Detectors
| Autor: | Chaoqi Chen, Zebiao Zheng, Xinghao Ding, Yue Huang, Qi Dou |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
FOS: Computer and information sciences
business.industry Computer science Computer Vision and Pattern Recognition (cs.CV) Feature extraction Computer Science - Computer Vision and Pattern Recognition Context (language use) Pattern recognition 02 engineering and technology 010501 environmental sciences Object (computer science) 01 natural sciences Object detection Discriminative model Feature (computer vision) 0202 electrical engineering electronic engineering information engineering Benchmark (computing) 020201 artificial intelligence & image processing Artificial intelligence business 0105 earth and related environmental sciences Interpolation |
| Zdroj: | CVPR |
| DOI: | 10.48550/arxiv.2003.06297 |
| Popis: | Recent advances in adaptive object detection have achieved compelling results in virtue of adversarial feature adaptation to mitigate the distributional shifts along the detection pipeline. Whilst adversarial adaptation significantly enhances the transferability of feature representations, the feature discriminability of object detectors remains less investigated. Moreover, transferability and discriminability may come at a contradiction in adversarial adaptation given the complex combinations of objects and the differentiated scene layouts between domains. In this paper, we propose a Hierarchical Transferability Calibration Network (HTCN) that hierarchically (local-region/image/instance) calibrates the transferability of feature representations for harmonizing transferability and discriminability. The proposed model consists of three components: (1) Importance Weighted Adversarial Training with input Interpolation (IWAT-I), which strengthens the global discriminability by re-weighting the interpolated image-level features; (2) Context-aware Instance-Level Alignment (CILA) module, which enhances the local discriminability by capturing the underlying complementary effect between the instance-level feature and the global context information for the instance-level feature alignment; (3) local feature masks that calibrate the local transferability to provide semantic guidance for the following discriminative pattern alignment. Experimental results show that HTCN significantly outperforms the state-of-the-art methods on benchmark datasets. Comment: Accepted by CVPR 2020 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|