GTNet: Generative Transfer Network for Zero-Shot Object Detection

• DOI: 10.1609/aaai.v34i07.6996
• Publication Date: 2020-06-19T08:20:33Z
• Authors: Shizhen Zhao; Changxin Gao; Yuanjie Shao; Lerenhan Li; Changqian Yu; Zhong Ji; Nong Sang
• Keywords: Feature (linguistics); Transfer of learning

We propose a Generative Transfer Network (GTNet) for zero-shot object detection (ZSD). GTNet consists of an Object Detection Module and Knowledge Module. The can learn large-scale seen domain knowledge. leverages feature synthesizer to generate unseen class features, which are applied train new classification layer the In order synthesize features each with both intra-class variance IoU variance, we design IoU-Aware Adversarial (IoUGAN) as synthesizer, be easily integrated into GTNet. Specifically, IoUGAN three unit models: Class Feature Generating Unit (CFU), Foreground (FFU), Background (BFU). CFU generates conditioned on semantic embeddings. FFU BFU add results CFU, yielding class-specific foreground background respectively. evaluate our method public datasets demonstrate that performs favorably against state-of-the-art ZSD approaches.