Contrastive Syn-to-Real Generalization
This addresses domain adaptation challenges for researchers and practitioners using synthetic data in label-scarce scenarios, but it is incremental as it builds on existing contrastive and pre-training methods.
The paper tackles the problem of poor generalization from synthetic to real data by proposing a contrastive framework that leverages ImageNet pre-training and promotes feature diversity, achieving state-of-the-art performance in zero-shot domain generalization.
Training on synthetic data can be beneficial for label or data-scarce scenarios. However, synthetically trained models often suffer from poor generalization in real domains due to domain gaps. In this work, we make a key observation that the diversity of the learned feature embeddings plays an important role in the generalization performance. To this end, we propose contrastive synthetic-to-real generalization (CSG), a novel framework that leverages the pre-trained ImageNet knowledge to prevent overfitting to the synthetic domain, while promoting the diversity of feature embeddings as an inductive bias to improve generalization. In addition, we enhance the proposed CSG framework with attentional pooling (A-pool) to let the model focus on semantically important regions and further improve its generalization. We demonstrate the effectiveness of CSG on various synthetic training tasks, exhibiting state-of-the-art performance on zero-shot domain generalization.