ReDDLE-Net: Reflectance Decomposition for Directional Light Estimation
The surfaces of real objects can visually appear to be glossy, matte, or anywhere in between, but essentially, they display varying degrees of diffuse and specular reflectance. Diffuse and specular reflectance provides different clues for light estimation. However, few methods simultaneously conside...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-09-01
|
| Series: | Photonics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2304-6732/9/9/656 |
| _version_ | 1797483478455943168 |
|---|---|
| author | Jiangxin Yang Binjie Ding Zewei He Gang Pan Yanpeng Cao Yanlong Cao Qian Zheng |
| author_facet | Jiangxin Yang Binjie Ding Zewei He Gang Pan Yanpeng Cao Yanlong Cao Qian Zheng |
| author_sort | Jiangxin Yang |
| collection | DOAJ |
| description | The surfaces of real objects can visually appear to be glossy, matte, or anywhere in between, but essentially, they display varying degrees of diffuse and specular reflectance. Diffuse and specular reflectance provides different clues for light estimation. However, few methods simultaneously consider the contributions of diffuse and specular reflectance for light estimation. To this end, we propose ReDDLE-Net, which performs <b>Re</b>flectance <b>D</b>ecomposition for <b>D</b>irectional <b>L</b>ight <b>E</b>stimation. The primary idea is to take advantage of diffuse and specular clues and adaptively balance the contributions of estimated diffuse and specular components for light estimation. Our method achieves a superior performance advantage over state-of-the-art directional light estimation methods on the DiLiGenT benchmark. Meanwhile, the proposed ReDDLE-Net can be combined with existing calibrated photometric stereo methods to handle uncalibrated photometric stereo tasks and achieve state-of-the-art performance. |
| first_indexed | 2024-03-09T22:48:32Z |
| format | Article |
| id | doaj.art-872e9706ebcc423f9b7b89285ab327c4 |
| institution | Directory Open Access Journal |
| issn | 2304-6732 |
| language | English |
| last_indexed | 2024-03-09T22:48:32Z |
| publishDate | 2022-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Photonics |
| spelling | doaj.art-872e9706ebcc423f9b7b89285ab327c42023-11-23T18:25:05ZengMDPI AGPhotonics2304-67322022-09-019965610.3390/photonics9090656ReDDLE-Net: Reflectance Decomposition for Directional Light EstimationJiangxin Yang0Binjie Ding1Zewei He2Gang Pan3Yanpeng Cao4Yanlong Cao5Qian Zheng6State Key Laboratory of Fluid Power and Mechatronic Systems, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, ChinaState Key Laboratory of Fluid Power and Mechatronic Systems, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, ChinaState Key Laboratory of Fluid Power and Mechatronic Systems, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, ChinaCollege of Computer Science and Technology, Zhejiang University, Hangzhou 310027, ChinaState Key Laboratory of Fluid Power and Mechatronic Systems, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, ChinaState Key Laboratory of Fluid Power and Mechatronic Systems, School of Mechanical Engineering, Zhejiang University, Hangzhou 310027, ChinaCollege of Computer Science and Technology, Zhejiang University, Hangzhou 310027, ChinaThe surfaces of real objects can visually appear to be glossy, matte, or anywhere in between, but essentially, they display varying degrees of diffuse and specular reflectance. Diffuse and specular reflectance provides different clues for light estimation. However, few methods simultaneously consider the contributions of diffuse and specular reflectance for light estimation. To this end, we propose ReDDLE-Net, which performs <b>Re</b>flectance <b>D</b>ecomposition for <b>D</b>irectional <b>L</b>ight <b>E</b>stimation. The primary idea is to take advantage of diffuse and specular clues and adaptively balance the contributions of estimated diffuse and specular components for light estimation. Our method achieves a superior performance advantage over state-of-the-art directional light estimation methods on the DiLiGenT benchmark. Meanwhile, the proposed ReDDLE-Net can be combined with existing calibrated photometric stereo methods to handle uncalibrated photometric stereo tasks and achieve state-of-the-art performance.https://www.mdpi.com/2304-6732/9/9/656uncalibrated photometric stereodiffuse reflectancespecular reflectanceconvolutional neural network |
| spellingShingle | Jiangxin Yang Binjie Ding Zewei He Gang Pan Yanpeng Cao Yanlong Cao Qian Zheng ReDDLE-Net: Reflectance Decomposition for Directional Light Estimation Photonics uncalibrated photometric stereo diffuse reflectance specular reflectance convolutional neural network |
| title | ReDDLE-Net: Reflectance Decomposition for Directional Light Estimation |
| title_full | ReDDLE-Net: Reflectance Decomposition for Directional Light Estimation |
| title_fullStr | ReDDLE-Net: Reflectance Decomposition for Directional Light Estimation |
| title_full_unstemmed | ReDDLE-Net: Reflectance Decomposition for Directional Light Estimation |
| title_short | ReDDLE-Net: Reflectance Decomposition for Directional Light Estimation |
| title_sort | reddle net reflectance decomposition for directional light estimation |
| topic | uncalibrated photometric stereo diffuse reflectance specular reflectance convolutional neural network |
| url | https://www.mdpi.com/2304-6732/9/9/656 |
| work_keys_str_mv | AT jiangxinyang reddlenetreflectancedecompositionfordirectionallightestimation AT binjieding reddlenetreflectancedecompositionfordirectionallightestimation AT zeweihe reddlenetreflectancedecompositionfordirectionallightestimation AT gangpan reddlenetreflectancedecompositionfordirectionallightestimation AT yanpengcao reddlenetreflectancedecompositionfordirectionallightestimation AT yanlongcao reddlenetreflectancedecompositionfordirectionallightestimation AT qianzheng reddlenetreflectancedecompositionfordirectionallightestimation |