Visual number sense for real-world scenes shared by deep neural networks and humans
Recently, visual number sense has been identified from deep neural networks (DNNs). However, whether DNNs have the same capacity for real-world scenes, rather than the simple geometric figures that are often tested, is unclear. In this study, we explore the number perception of scenes using AlexNet...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-08-01
|
| Series: | Heliyon |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2405844023057250 |
| _version_ | 1797732972042911744 |
|---|---|
| author | Wu Wencheng Yingxi Ge Zhentao Zuo Lin Chen Xu Qin Liu Zuxiang |
| author_facet | Wu Wencheng Yingxi Ge Zhentao Zuo Lin Chen Xu Qin Liu Zuxiang |
| author_sort | Wu Wencheng |
| collection | DOAJ |
| description | Recently, visual number sense has been identified from deep neural networks (DNNs). However, whether DNNs have the same capacity for real-world scenes, rather than the simple geometric figures that are often tested, is unclear. In this study, we explore the number perception of scenes using AlexNet and find that numerosity can be represented by the pattern of group activation of the category layer units. The global activation of these units increases with the number of objects in the scene, and the variations in their activation decrease accordingly. By decoding the numerosity from this pattern, we reveal that the embedding coefficient of a scene determines the likelihood of potential objects to contribute to numerical perception. This was demonstrated by the more optimized performance for pictures with relatively high embedding coefficients in both DNNs and humans. This study for the first time shows that a distinct feature in visual environments, revealed by DNNs, can modulate human perception, supported by a group-coding mechanism. |
| first_indexed | 2024-03-12T12:22:14Z |
| format | Article |
| id | doaj.art-0331c471ae1d40ffb4e8c5d28050d76a |
| institution | Directory Open Access Journal |
| issn | 2405-8440 |
| language | English |
| last_indexed | 2024-03-12T12:22:14Z |
| publishDate | 2023-08-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Heliyon |
| spelling | doaj.art-0331c471ae1d40ffb4e8c5d28050d76a2023-08-30T05:51:36ZengElsevierHeliyon2405-84402023-08-0198e18517Visual number sense for real-world scenes shared by deep neural networks and humansWu Wencheng0Yingxi Ge1Zhentao Zuo2Lin Chen3Xu Qin4Liu Zuxiang5AHU-IAI AI Joint Laboratory, Anhui University, Hefei, 230601, China; Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, 230088, ChinaState Key Laboratory of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences, 15 Datun Road, Beijing, 100101, China; CAS Center for Excellence in Brain Science and Intelligence Technology, China; University of Chinese Academy of Sciences, 19A Yuquan Road, Beijing, 100049, ChinaInstitute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, 230088, China; State Key Laboratory of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences, 15 Datun Road, Beijing, 100101, China; CAS Center for Excellence in Brain Science and Intelligence Technology, China; University of Chinese Academy of Sciences, 19A Yuquan Road, Beijing, 100049, ChinaInstitute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, 230088, China; State Key Laboratory of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences, 15 Datun Road, Beijing, 100101, China; CAS Center for Excellence in Brain Science and Intelligence Technology, China; University of Chinese Academy of Sciences, 19A Yuquan Road, Beijing, 100049, ChinaKey Laboratory of Intelligent Computing and Signal Processing of Ministry of Education, Hefei, 230601, China; Anhui Provincial Key Laboratory of Multimodal Cognitive Computation, Anhui University, Hefei, 230601, China; School of Computer Science and Technology, Anhui University, Hefei 230601, China; Corresponding author. School of Computer Science and Technology, Anhui University, Hefei 230601, China.Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, 230088, China; State Key Laboratory of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences, 15 Datun Road, Beijing, 100101, China; CAS Center for Excellence in Brain Science and Intelligence Technology, China; University of Chinese Academy of Sciences, 19A Yuquan Road, Beijing, 100049, China; Corresponding author. State Key Laboratory of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences, 15 Datun Road, Beijing 100101, China.Recently, visual number sense has been identified from deep neural networks (DNNs). However, whether DNNs have the same capacity for real-world scenes, rather than the simple geometric figures that are often tested, is unclear. In this study, we explore the number perception of scenes using AlexNet and find that numerosity can be represented by the pattern of group activation of the category layer units. The global activation of these units increases with the number of objects in the scene, and the variations in their activation decrease accordingly. By decoding the numerosity from this pattern, we reveal that the embedding coefficient of a scene determines the likelihood of potential objects to contribute to numerical perception. This was demonstrated by the more optimized performance for pictures with relatively high embedding coefficients in both DNNs and humans. This study for the first time shows that a distinct feature in visual environments, revealed by DNNs, can modulate human perception, supported by a group-coding mechanism.http://www.sciencedirect.com/science/article/pii/S2405844023057250Number senseDeep neural networkReal-world sceneGroup codingEmbedded representation |
| spellingShingle | Wu Wencheng Yingxi Ge Zhentao Zuo Lin Chen Xu Qin Liu Zuxiang Visual number sense for real-world scenes shared by deep neural networks and humans Heliyon Number sense Deep neural network Real-world scene Group coding Embedded representation |
| title | Visual number sense for real-world scenes shared by deep neural networks and humans |
| title_full | Visual number sense for real-world scenes shared by deep neural networks and humans |
| title_fullStr | Visual number sense for real-world scenes shared by deep neural networks and humans |
| title_full_unstemmed | Visual number sense for real-world scenes shared by deep neural networks and humans |
| title_short | Visual number sense for real-world scenes shared by deep neural networks and humans |
| title_sort | visual number sense for real world scenes shared by deep neural networks and humans |
| topic | Number sense Deep neural network Real-world scene Group coding Embedded representation |
| url | http://www.sciencedirect.com/science/article/pii/S2405844023057250 |
| work_keys_str_mv | AT wuwencheng visualnumbersenseforrealworldscenessharedbydeepneuralnetworksandhumans AT yingxige visualnumbersenseforrealworldscenessharedbydeepneuralnetworksandhumans AT zhentaozuo visualnumbersenseforrealworldscenessharedbydeepneuralnetworksandhumans AT linchen visualnumbersenseforrealworldscenessharedbydeepneuralnetworksandhumans AT xuqin visualnumbersenseforrealworldscenessharedbydeepneuralnetworksandhumans AT liuzuxiang visualnumbersenseforrealworldscenessharedbydeepneuralnetworksandhumans |