College of Computing and Mathematics
Information & Computer Science Department
Presents an Online Seminar
"Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation"
Date: Monday, 3rd October 2022
Time: 02:30 pm – 3:45 pm
Location: Bldg. 59, Room 1005
Mr. M Faisal Nurnoby
PhD, ICS Dept.
King Fahd University of Petroleum & Minerals.
Spatial pyramid pooling module or encode-decoder structure are used in deep neural networks for semantic segmentation task. The former networks are able to encode multi-scale contextual information by probing the incoming features with filters or pooling operations at multiple rates and multiple effective fields-of-view, while the latter networks can capture sharper object boundaries by gradually recovering the spatial information. In this work, we propose to combine the advantages from both methods. Specifically, our proposed model, DeepLabv3+, extends DeepLabv3 by adding a simple yet effective decoder module to refine the segmentation results especially along object boundaries. We further explore the Xception model and apply the depth-wise separable convolution to both Atrous Spatial Pyramid Pooling and decoder modules, resulting in a faster and stronger encoder-decoder network. We demonstrate the effectiveness of the proposed model on PASCAL VOC 2012 and Cityscapes datasets, achieving the test set performance of 89% and 82.1% without any post-processing. Our paper is accompanied with a publicly available reference implementation of the proposed models in Tensorflow.
Short bio of presenter:
M Faisal Nurnoby, is a full-time PhD student at KFUPM. His major is computer science. His research interests are computer vision, semantic segmentation, human-computer interaction (HCI), and deep learning. Previously he had been teaching at North South University on a full-time basis until 2017. Also, he has a track record of developing software for both large local and multinational companies.
All faculty, researchers and graduate students are invited to attend.
Information & Computer Science Department, College of Computing and Mathematics
Telephone: +966 (13) 860 2175, Email: email@example.com, Website: www.kfupm.edu.sa/departments/ics/
Copyright © 2014 King Fahd University of Petroleum & Minerals