Towards High Performance Video Object Detection for Mobiles
Published in
arXiv.org(2018)External Links:
TL
TL;DR
A light weight network architecture for video object detection on mobiles, with a very small network for establishing correspondence across frames, and a flow-guided GRU module designed to effectively aggregate features on key frames.
Abstract
Despite the recent success of video object detection on Desktop GPUs, its architecture is still far too heavy for mobiles. It is also unclear whether the key principles of sparse feature propagation and multi-frame feature aggregation apply at very limited computational resources. In this paper, we present a light weight network architecture for video object detection on mobiles. Light weight image object detector is applied on sparse key frames. A very small network, Light Flow, is designed for establishing correspondence across frames. A flow-guided GRU module is designed to effectively aggregate features on key frames. For non-key frames, sparse feature propagation is performed. The whole network can be trained end-to-end. The proposed system achieves 60.2% mAP score at speed of 25.6 fps on mobiles (e.g., HuaWei Mate 8).
References54 items
1
Tiny SSD: A Tiny Single-Shot Detection Deep Convolutional Neural Network for Real-Time Embedded Object Detection
2
Inverted Residuals and Linear Bottlenecks: Mobile Networks for Classification, Detection and Segmentation
3
MobileNetV2: Inverted Residuals and Linear Bottlenecks
4
Towards High Performance Video Object Detection
5