E2VTS: Energy-Efficient Video Text Spotting from Unmanned Aerial Vehicles

Unmanned Aerial Vehicles (UAVs) based video text spot-ting has been extensively used in civil and military domains. UAV's limited battery capacity motivates us to develop an energy-efficient video text spotting solution. In this paper, we first revisit RCNN's crop & resize training str...

Full description

Saved in:
Bibliographic Details
Main Authors: Hu, Zhenyu, Pi, Pengcheng, Wu, Zhenyu, Xue, Yunhe, Shen, Jiayi, Tan, Jianchao, Lian, Xiangru, Wang, Zhangyang, Liu, Ji
Format: Conference Proceeding
Language:English
Subjects:
Energy efficiency
Heuristic algorithms
Image edge detection
Quantization (signal)
Redundancy
Routing
Training
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Unmanned Aerial Vehicles (UAVs) based video text spot-ting has been extensively used in civil and military domains. UAV's limited battery capacity motivates us to develop an energy-efficient video text spotting solution. In this paper, we first revisit RCNN's crop & resize training strategy and empirically find that it outperforms aligned RoI sampling on a real-world video text dataset captured by UAV. To re-duce energy consumption, we further propose a multi-stage image processor that takes videos' redundancy, continuity, and mixed degradation into account. The model is pruned and quantized before deployed on Raspberry Pi. Our pro-posed energy-efficient video text spotting solution, dubbed as E 2 V T S, outperforms all previous methods by achieving a competitive tradeoff between energy efficiency and performance. All our codes and pre-trained models are available at https://github.com/wuzhenyusjtu/LPCVC20-VideoTextSpotting.
ISSN:2160-7516
DOI:10.1109/CVPRW53098.2021.00101