Creating Lightweight Object Detectors with Model Compression for Deployment on Edge Devices
This addresses the need for efficient object detection on resource-constrained edge devices, presenting an incremental improvement over existing methods.
The paper tackles the problem of deploying object detectors on edge devices by proposing a model compression pipeline, achieving a lightweight SSD-300 with 16.3MB size, 2.31G FLOPS, and 71.2 mAP, outperforming SSD-300-MobileNet.
To achieve lightweight object detectors for deployment on the edge devices, an effective model compression pipeline is proposed in this paper. The compression pipeline consists of automatic channel pruning for the backbone, fixed channel deletion for the branch layers and knowledge distillation for the guidance learning. As results, the Resnet50-v1d is auto-pruned and fine-tuned on ImageNet to attain a compact base model as the backbone of object detector. Then, lightweight object detectors are implemented with proposed compression pipeline. For instance, the SSD-300 with model size=16.3MB, FLOPS=2.31G, and mAP=71.2 is created, revealing a better result than SSD-300-MobileNet.