SIMULATION-AND-MINING: TOWARDS ACCURATE SOURCE-FREE UNSUPERVISED DOMAIN ADAPTIVE OBJECT DETECTION

Vanilla unsupervised domain adaptive (UDA) object detection typically requires the labeled source data for joint-training with the unlabeled target data, which is usually unavailable in real-world scenarios due to data privacy, leading to source data-free UDA object detection. Herein, we first conduct evaluations to analyze the phenomenon of detecting degradation in hard samples (e.g. small-scale or occluded objects ) after domain adaption, terming as domain generalization differentiation. To this end, we then revisit the existing self-training method, which is of great challenge to deal with the abundant false negatives (hard samples). Assume that true positives (easy samples) labeled by the source model can be exploited as supervision cues. UDA is finally modeled into an unsupervised false negatives mining problem. Thus, we propose a Simulation-and-Mining (S&M) framework, which simulates false negatives by augmenting true positives and mines back false negatives alternatively and iteratively. Experimental results show the superiority of our S&M framework.