Seeing is Worse than Believing: Reading People’s Minds Better than Computer-Vision Methods Recognize Actions
We had human subjects perform a one-out-of-six class action recognition task from video stimuli while undergoing functional magnetic resonance imaging (fMRI). Support-vector machines (SVMs) were trained on the recovered brain scans to classify actions observed during imaging, yielding average classi...
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Technical Report |
| Language: | en_US |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/100176 |
| _version_ | 1811078112967917568 |
|---|---|
| author | Barbu, Andrei Barrett, Daniel P. Chen, Wei Narayanaswamy, Siddharth Xiong, Caiming Corso, Jason J. Fellbaum, Christiane D. Hanson, Catherine Hanson, Stephen Jose Helie, Sebastien Malaia, Evguenia Pearlmutter, Barak A. Siskind, Jeffrey Mark Talavage, Thomas Michael Wilbur, Ronnie B. |
| author_facet | Barbu, Andrei Barrett, Daniel P. Chen, Wei Narayanaswamy, Siddharth Xiong, Caiming Corso, Jason J. Fellbaum, Christiane D. Hanson, Catherine Hanson, Stephen Jose Helie, Sebastien Malaia, Evguenia Pearlmutter, Barak A. Siskind, Jeffrey Mark Talavage, Thomas Michael Wilbur, Ronnie B. |
| author_sort | Barbu, Andrei |
| collection | MIT |
| description | We had human subjects perform a one-out-of-six class action recognition task from video stimuli while undergoing functional magnetic resonance imaging (fMRI). Support-vector machines (SVMs) were trained on the recovered brain scans to classify actions observed during imaging, yielding average classification accuracy of 69.73% when tested on scans from the same subject and of 34.80% when tested on scans from different subjects. An apples-to-apples comparison was performed with all publicly available software that implements state-of-the-art action recognition on the same video corpus with the same cross-validation regimen and same partitioning into training and test sets, yielding classification accuracies between 31.25% and 52.34%. This indicates that one can read people’s minds better than state-of-the-art computer-vision methods can perform action recognition. |
| first_indexed | 2024-09-23T10:53:36Z |
| format | Technical Report |
| id | mit-1721.1/100176 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T10:53:36Z |
| publishDate | 2015 |
| record_format | dspace |
| spelling | mit-1721.1/1001762019-04-12T07:33:55Z Seeing is Worse than Believing: Reading People’s Minds Better than Computer-Vision Methods Recognize Actions Barbu, Andrei Barrett, Daniel P. Chen, Wei Narayanaswamy, Siddharth Xiong, Caiming Corso, Jason J. Fellbaum, Christiane D. Hanson, Catherine Hanson, Stephen Jose Helie, Sebastien Malaia, Evguenia Pearlmutter, Barak A. Siskind, Jeffrey Mark Talavage, Thomas Michael Wilbur, Ronnie B. Object Recognition Vision Support-Vector Machines (SVMs) We had human subjects perform a one-out-of-six class action recognition task from video stimuli while undergoing functional magnetic resonance imaging (fMRI). Support-vector machines (SVMs) were trained on the recovered brain scans to classify actions observed during imaging, yielding average classification accuracy of 69.73% when tested on scans from the same subject and of 34.80% when tested on scans from different subjects. An apples-to-apples comparison was performed with all publicly available software that implements state-of-the-art action recognition on the same video corpus with the same cross-validation regimen and same partitioning into training and test sets, yielding classification accuracies between 31.25% and 52.34%. This indicates that one can read people’s minds better than state-of-the-art computer-vision methods can perform action recognition. This work was supported, in part, by the Center for Brains, Minds and Machines (CBMM), funded by NSF STC award CCF - 1231216. AB, DPB, NS, and JMS were supported, in part, by Army Research Laboratory (ARL) Cooperative Agreement W911NF-10-2-0060, AB, in part, by the Center forBrains, Minds and Machines (CBMM), funded by NSF STC award CCF-1231216, WC, CX, and JJC, in part, by ARL Cooperative Agreement W911NF-10-2-0062 and NSF CAREER grant IIS-0845282, CDF, in part, by NSF grant CNS-0855157, CH and SJH, in part, by the McDonnell Foundation, and BAP, in part, by Science Foundation Ireland grant 09/IN.1/I2637. 2015-12-10T19:07:10Z 2015-12-10T19:07:10Z 2015-12-10 Technical Report Working Paper Other http://hdl.handle.net/1721.1/100176 en_US CBMM Memo Series;012 Attribution-NonCommercial 3.0 United States http://creativecommons.org/licenses/by-nc/3.0/us/ application/pdf |
| spellingShingle | Object Recognition Vision Support-Vector Machines (SVMs) Barbu, Andrei Barrett, Daniel P. Chen, Wei Narayanaswamy, Siddharth Xiong, Caiming Corso, Jason J. Fellbaum, Christiane D. Hanson, Catherine Hanson, Stephen Jose Helie, Sebastien Malaia, Evguenia Pearlmutter, Barak A. Siskind, Jeffrey Mark Talavage, Thomas Michael Wilbur, Ronnie B. Seeing is Worse than Believing: Reading People’s Minds Better than Computer-Vision Methods Recognize Actions |
| title | Seeing is Worse than Believing: Reading People’s Minds Better than Computer-Vision Methods Recognize Actions |
| title_full | Seeing is Worse than Believing: Reading People’s Minds Better than Computer-Vision Methods Recognize Actions |
| title_fullStr | Seeing is Worse than Believing: Reading People’s Minds Better than Computer-Vision Methods Recognize Actions |
| title_full_unstemmed | Seeing is Worse than Believing: Reading People’s Minds Better than Computer-Vision Methods Recognize Actions |
| title_short | Seeing is Worse than Believing: Reading People’s Minds Better than Computer-Vision Methods Recognize Actions |
| title_sort | seeing is worse than believing reading people s minds better than computer vision methods recognize actions |
| topic | Object Recognition Vision Support-Vector Machines (SVMs) |
| url | http://hdl.handle.net/1721.1/100176 |
| work_keys_str_mv | AT barbuandrei seeingisworsethanbelievingreadingpeoplesmindsbetterthancomputervisionmethodsrecognizeactions AT barrettdanielp seeingisworsethanbelievingreadingpeoplesmindsbetterthancomputervisionmethodsrecognizeactions AT chenwei seeingisworsethanbelievingreadingpeoplesmindsbetterthancomputervisionmethodsrecognizeactions AT narayanaswamysiddharth seeingisworsethanbelievingreadingpeoplesmindsbetterthancomputervisionmethodsrecognizeactions AT xiongcaiming seeingisworsethanbelievingreadingpeoplesmindsbetterthancomputervisionmethodsrecognizeactions AT corsojasonj seeingisworsethanbelievingreadingpeoplesmindsbetterthancomputervisionmethodsrecognizeactions AT fellbaumchristianed seeingisworsethanbelievingreadingpeoplesmindsbetterthancomputervisionmethodsrecognizeactions AT hansoncatherine seeingisworsethanbelievingreadingpeoplesmindsbetterthancomputervisionmethodsrecognizeactions AT hansonstephenjose seeingisworsethanbelievingreadingpeoplesmindsbetterthancomputervisionmethodsrecognizeactions AT heliesebastien seeingisworsethanbelievingreadingpeoplesmindsbetterthancomputervisionmethodsrecognizeactions AT malaiaevguenia seeingisworsethanbelievingreadingpeoplesmindsbetterthancomputervisionmethodsrecognizeactions AT pearlmutterbaraka seeingisworsethanbelievingreadingpeoplesmindsbetterthancomputervisionmethodsrecognizeactions AT siskindjeffreymark seeingisworsethanbelievingreadingpeoplesmindsbetterthancomputervisionmethodsrecognizeactions AT talavagethomasmichael seeingisworsethanbelievingreadingpeoplesmindsbetterthancomputervisionmethodsrecognizeactions AT wilburronnieb seeingisworsethanbelievingreadingpeoplesmindsbetterthancomputervisionmethodsrecognizeactions |