Changfeng Chen

Changfeng Chen

Auto-generated profile, based on publicly available information

Changfeng Chen

Southwest China Research Institute of Electronic Equipment, China, China

Research Focus & Metrics

Representative Publications

View All Publications
23 February 2024
Verified
Conference Paper
Open access
XZXiaoliang Zhang
Xiaoliang Zhang
YLYuankun Liu
Yuankun Liu
MLMao Li
Mao Li
YZYuqing Zhang
Yuqing Zhang
CCChangfeng Chen
Changfeng Chen
XZXiantao Zhou
Xiantao Zhou
Instance segmentation is an emerging and challenging task. A lot of previous methods have focused on this field and made great progress. However, complex background and mutual occlusion between instances still hinder the development of instance segmentation. In this paper, we aim to solve foreground confusion for instance segmentation. Based on Mask R-CNN, we first add an additional mask branch, which is parallel to the existing mask branch and expected to predict non-target foreground pixels within the predicted bounding box. To differentiate the two mask branches, the original mask branch is called Target Mask Branch while the additional mask branch is called Non-Target Mask Branch. Besides, in order to enhance the feature interaction and differences between the target mask branch and the non-target mask branch, we design a foreground confusion processing module to connect the dual mask branches. In this way, our method can distinguish the target instance and non-target foreground pixels within the corresponding predicted bounding box effectively, and obtain more accurate segmentation masks. Furthermore, a feature distinguish module is proposed to enhance the differences between features from the dual mask branches. We demonstrate our method's competitiveness on MS COCO dataset, which outperforms previous state-of-the-art instance segmentation methods. In addition, we also add our method into other instance segmentation methods, and the experiment results show our method still can obtain stable gains in performance.

12 April 2024
Verified
Conference Paper
Open access
YZYang Zhang
Yang Zhang
CCChangfeng Chen
Changfeng Chen
YZYuqing Zhang
Yuqing Zhang
QLQiangbin Li
Qiangbin Li
HWHao Wang
Hao Wang
MZMeng Zou
Meng Zou
In recent days, wireless sensors and communication signals are not limited to signal interaction and transmission, but more application possibilities have emerged, such as wireless environment awareness, human condition monitoring, etc. In this paper, we take into account the received signal strength indication information of commercial wireless sensor WiFi signal to obtain two-dimensional profile imaging results of human body and obstacle structures. More specifically, we establish an approximate electromagnetic propagation model between transmitting and receiving wireless sensors, which reveals the attenuation characteristics of radio waves passing through obstacles. Then, a modified statistical iterative imaging method based on the initial imaging matrix of filtered back projection tomography is proposed. Finally, the experimental results validate the effectiveness of the proposed approximate electromagnetic propagation model and imaging method.