Learning person-person interaction in collective activity recognition
- Submitting institution
-
University of Dundee
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 28402999
- Type
- D - Journal article
- DOI
-
10.1109/TIP.2015.2409564
- Title of journal
- IEEE Transactions on Image Processing
- Article number
- 7055886
- First page
- 1905
- Volume
- 24
- Issue
- 6
- ISSN
- 1057-7149
- Open access status
- Out of scope for open access requirements
- Month of publication
- March
- Year of publication
- 2015
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
2
- Research group(s)
-
-
- Citation count
- 24
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Stressing the importance of person-person interactions, this paper proposes a learning-based approach for collective activity recognition capable of modelling class-specific interactions that achieves state-of-the-art performance on two benchmarking datasets. The argument and algorithm have inspired and enabled the design of a series of methods modelling interactions for collective activity recognition, including Concurrent Memory (Shu et al., IEEE Tran. PAMI, 2019) involving Tencent, Graph LSTM (Tang et al, IEEE T-PAMI, 2019), and Tang et al. (IEEE Trans. Image Processing, 2019) involving Horizon Robotics Inc.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -