Cross Input Neighbourhood Difference for Re-identification of Occupational Humans
- Resource Type
- Conference
- Authors
- Mohammad, Agha Saad; Saleem, Summra; Dilawari, Aniqa; Ghani Khan, Muhammad Usman
- Source
- 2019 22nd International Multitopic Conference (INMIC) Multitopic Conference (INMIC), 2019 22nd International. :1-6 Nov, 2019
- Subject
- Communication, Networking and Broadcast Technologies
Computing and Processing
General Topics for Engineers
Power, Energy and Industry Applications
Robotics and Control Systems
Signal Processing and Analysis
Feature extraction
Security
Task analysis
Clothing
Target tracking
Surveillance
CNN
Re-identification
Classification
Human detection
Apparel Identification
- Language
- ISSN
- 2049-3630
In this research study, we have proposed an appearance-based classifier to classify human corresponding to their attire and clothing. Personality re-identification proposes to address the issue of keeping track of persons. Our proposed methodology presents the advance work on both CUHK03 and CUHK01 dataset, and bypass over-fitting problem by artificially enlarging the dataset using label-preserving transformations. The model is fine-tunned on a small amount of target dataset, which achieved results equal to the state-of-the-art. Deep convolutional architecture has been used with Cross input Neighborhood difference to solve the problem of re-identification through robustness and positional differences.