Context Unaware Knowledge Distillation for Image Retrieval
This work addresses efficiency issues in image retrieval for applications requiring compact models, though it is incremental as it builds on existing knowledge distillation techniques.
The paper tackles the computational complexity of data-dependent hashing methods by proposing context unaware knowledge distillation, which allows using a teacher model without fine-tuning it on the target context, and introduces an efficient student architecture, achieving a promising trade-off between retrieval results and efficiency in frameworks like DCH and CSQ.
Existing data-dependent hashing methods use large backbone networks with millions of parameters and are computationally complex. Existing knowledge distillation methods use logits and other features of the deep (teacher) model and as knowledge for the compact (student) model, which requires the teacher's network to be fine-tuned on the context in parallel with the student model on the context. Training teacher on the target context requires more time and computational resources. In this paper, we propose context unaware knowledge distillation that uses the knowledge of the teacher model without fine-tuning it on the target context. We also propose a new efficient student model architecture for knowledge distillation. The proposed approach follows a two-step process. The first step involves pre-training the student model with the help of context unaware knowledge distillation from the teacher model. The second step involves fine-tuning the student model on the context of image retrieval. In order to show the efficacy of the proposed approach, we compare the retrieval results, no. of parameters and no. of operations of the student models with the teacher models under different retrieval frameworks, including deep cauchy hashing (DCH) and central similarity quantization (CSQ). The experimental results confirm that the proposed approach provides a promising trade-off between the retrieval results and efficiency. The code used in this paper is released publicly at \url{https://github.com/satoru2001/CUKDFIR}.