BADANet: Boundary Aware Dilated Attention Network for Face Parsing
Over the past few years, deep learning techniques have revolutionized the field of face parsing by utilizing massive datasets to generate high-level features and achieve outstanding performance. Usually, these techniques involve Convolutional Neural Networks (CNNs) to derive features from the input...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10264074/ |
| _version_ | 1797663087004745728 |
|---|---|
| author | S. Raghavendra S. K. Abhilash Venu Madhav Nookala N. N. Srinidhi N. D. Adesh |
| author_facet | S. Raghavendra S. K. Abhilash Venu Madhav Nookala N. N. Srinidhi N. D. Adesh |
| author_sort | S. Raghavendra |
| collection | DOAJ |
| description | Over the past few years, deep learning techniques have revolutionized the field of face parsing by utilizing massive datasets to generate high-level features and achieve outstanding performance. Usually, these techniques involve Convolutional Neural Networks (CNNs) to derive features from the input image and then a decoder network to forecast the semantic labels for every pixel. Even with complex deep convolutional neural networks (DCNN), incorrect parsing leads to a semantic gap between identical features, especially at boundary levels. Face parsing has encountered additional challenges due to the inclusion of pose, illumination and facial expressions. In order to address these issues, a Boundary Aware Dilated Attention Network (BADANet) is introduced which explores the use of multi-scale techniques to improve the accuracy and robustness of the per-pixel frames. BADANet’s dilated attention module has been combined with a lightweight backbone to achieve exceptional results on LaPa, CelebAMask-HQ, and iBugMask. Extensive evaluations of the proposed method demonstrate its performance to be on par with various state-of-the-art techniques. The pipeline for the BADANet is available at <uri>https://github.com/abhigoku10/BADANet</uri> |
| first_indexed | 2024-03-11T19:09:27Z |
| format | Article |
| id | doaj.art-bb9d28ace5194bf9a79c0fea8c08e935 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-11T19:09:27Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-bb9d28ace5194bf9a79c0fea8c08e9352023-10-09T23:01:18ZengIEEEIEEE Access2169-35362023-01-011110674910675910.1109/ACCESS.2023.331956110264074BADANet: Boundary Aware Dilated Attention Network for Face ParsingS. Raghavendra0https://orcid.org/0000-0003-2733-3916S. K. Abhilash1https://orcid.org/0000-0002-1119-4782Venu Madhav Nookala2https://orcid.org/0000-0002-0078-5050N. N. Srinidhi3https://orcid.org/0000-0002-2865-1185N. D. Adesh4https://orcid.org/0000-0002-9750-113XDepartment of Information and Communication Technology, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, IndiaPathPartner Technology Pvt. Ltd., Bengaluru, IndiaPathPartner Technology Pvt. Ltd., Bengaluru, IndiaDepartment of Computer Science and Engineering, Manipal Institute of Technology Bengaluru, Manipal Academy of Higher Education, Manipal, IndiaDepartment of Information and Communication Technology, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, IndiaOver the past few years, deep learning techniques have revolutionized the field of face parsing by utilizing massive datasets to generate high-level features and achieve outstanding performance. Usually, these techniques involve Convolutional Neural Networks (CNNs) to derive features from the input image and then a decoder network to forecast the semantic labels for every pixel. Even with complex deep convolutional neural networks (DCNN), incorrect parsing leads to a semantic gap between identical features, especially at boundary levels. Face parsing has encountered additional challenges due to the inclusion of pose, illumination and facial expressions. In order to address these issues, a Boundary Aware Dilated Attention Network (BADANet) is introduced which explores the use of multi-scale techniques to improve the accuracy and robustness of the per-pixel frames. BADANet’s dilated attention module has been combined with a lightweight backbone to achieve exceptional results on LaPa, CelebAMask-HQ, and iBugMask. Extensive evaluations of the proposed method demonstrate its performance to be on par with various state-of-the-art techniques. The pipeline for the BADANet is available at <uri>https://github.com/abhigoku10/BADANet</uri>https://ieeexplore.ieee.org/document/10264074/BADANetbinary cross entropyCNNface parsingindustry innovation |
| spellingShingle | S. Raghavendra S. K. Abhilash Venu Madhav Nookala N. N. Srinidhi N. D. Adesh BADANet: Boundary Aware Dilated Attention Network for Face Parsing IEEE Access BADANet binary cross entropy CNN face parsing industry innovation |
| title | BADANet: Boundary Aware Dilated Attention Network for Face Parsing |
| title_full | BADANet: Boundary Aware Dilated Attention Network for Face Parsing |
| title_fullStr | BADANet: Boundary Aware Dilated Attention Network for Face Parsing |
| title_full_unstemmed | BADANet: Boundary Aware Dilated Attention Network for Face Parsing |
| title_short | BADANet: Boundary Aware Dilated Attention Network for Face Parsing |
| title_sort | badanet boundary aware dilated attention network for face parsing |
| topic | BADANet binary cross entropy CNN face parsing industry innovation |
| url | https://ieeexplore.ieee.org/document/10264074/ |
| work_keys_str_mv | AT sraghavendra badanetboundaryawaredilatedattentionnetworkforfaceparsing AT skabhilash badanetboundaryawaredilatedattentionnetworkforfaceparsing AT venumadhavnookala badanetboundaryawaredilatedattentionnetworkforfaceparsing AT nnsrinidhi badanetboundaryawaredilatedattentionnetworkforfaceparsing AT ndadesh badanetboundaryawaredilatedattentionnetworkforfaceparsing |