One-Shot Knowledge Transfer for Scalable Person Re-Identification
This work addresses the challenge of scalable model deployment for edge computing in person re-identification, offering a more efficient solution for scenarios with varying resource constraints, though it is incremental in the field of model compression.
The paper tackles the problem of repetitive computations in compressing person re-identification models for edge devices by proposing OSKT, a one-shot knowledge transfer method that uses a weight chain to generate models of varying sizes without additional computation, achieving significant performance improvements over state-of-the-art compression methods.
Edge computing in person re-identification (ReID) is crucial for reducing the load on central cloud servers and ensuring user privacy. Conventional compression methods for obtaining compact models require computations for each individual student model. When multiple models of varying sizes are needed to accommodate different resource conditions, this leads to repetitive and cumbersome computations. To address this challenge, we propose a novel knowledge inheritance approach named OSKT (One-Shot Knowledge Transfer), which consolidates the knowledge of the teacher model into an intermediate carrier called a weight chain. When a downstream scenario demands a model that meets specific resource constraints, this weight chain can be expanded to the target model size without additional computation. OSKT significantly outperforms state-of-the-art compression methods, with the added advantage of one-time knowledge transfer that eliminates the need for frequent computations for each target model.