Sort:
Open Access Research Article Issue
Global video object segmentation with spatial constraint module
Computational Visual Media 2023, 9 (2): 385-400
Published: 03 January 2023
Downloads:79

We present a lightweight and efficient semi-supervised video object segmentation network based on the space-time memory framework. To some extent, our method solves the two difficulties encountered in traditional video object segmentation: one is that the single frame calculation time is too long, and the other is that the current frame’s segmentation should use more information from past frames. The algorithm uses a global context (GC) module to achieve high-performance, real-time segmentation. The GC module can effectively integrate multi-frame image information without increased memory and can process each frame in real time. Moreover, the prediction mask of the previous frame is helpful for the segmentation of the current frame, so we input it into a spatial constraint module (SCM), which constrains the areas of segments in the current frame. The SCM effectively alleviates mismatching of similar targets yet consumes few additional resources. We added a refinement module to the decoder to improve boundary segmentation. Our model achieves state-of-the-art results on various datasets, scoring 80.1% on YouTube-VOS 2018 and a 𝒥& score of 78.0% on DAVIS 2017, while taking 0.05 s per frame on the DAVIS 2016 validation dataset.

Research Article Issue
A self-powered piezoelectret sensor based on foamed plastic garbage for monitoring human motions
Nano Research 2023, 16 (1): 1269-1276
Published: 30 August 2022
Downloads:76

Constructing piezoelectret based on foamed plastic garbage is an advisable strategy for obtaining self-powered flexible electromechanical sensors with good performances. Herein, a self-powered piezoelectret sensor with basic material of low density polyethylene (LDPE) foamed plastic garbage is proposed, with characteristics of easy fabrication, excellent flexibility, and high equivalent piezoelectric coefficient d33 value up to ~ 1,100 pC/N. The output stability is verified by continuously stimulating a sensor for ~ 180,000 cycles under low and high applied pressure, and the variations of peak outputs are less than 5.5%. Applications for measuring low- and high-pressure signals from human body are achieved. Assembled with a wristband, a sensor is demonstrated for detecting the human pulse waves. Moreover, real time human sitting information is wirelessly monitored with a smart chair based on 4 pixels sensors array.

total 2