- Home
- Research
- Research Area
- Visual Image Processing
Visual Image Processing
(a) to cultivate the awareness of the concept, benefits, applications and impact of VIP research to students, industry and community
(b) to provide VIP skill-oriented training for students, digital executives, system architects/engineers, and software developers
(c) to support local industries research and innovative developments in the field of VIP
(d) to become the point of reference for the implementation of the VIP to explore the potential use of digital image data to help improve the quality of social well-being, business process, decisions and services
1. Education
-
Effects of Visual Granularity in Visual Mnemonics on the learning of reactions in organic chemistry (Ts Dr Lim Phei Chin)
2. Health
-
A New Aritificial Neural Network Model to Predict Left Ventricle Remodeling for Clinical Decision and Treatment Strategies (Associate Professor Ts Dr Dayang Nurfatimah Awg Iskandar)
-
Automated Cardiac MRI Oedema System (ACES) (Associate Professor Ts Dr Dayang Nurfatimah Awg Iskandar)
-
Clinically Applicable Machine Learning Model for Classification of COVID-19 Abnormalities using Chest Radiographs (Dr Chai Soo See)
- Physiotherapy movement Recognition and Classification by Deep Learning Approach
-
A novel User Interface/Experience (UI/UX) for MySejahtera Usage Among Senior Users in Rural Sarawak (Samarahan, Simunjan and Betong) : Self-Monitoring Module (Mr Jonathan Sidi)
-
MyAsthma: From Axiomatization to a new Explanable-Artificial Intelligence-Based Risk Management Model with Progressive Web Application for care of Bronchial Asthma (Ts Dr Lim Phei Chin)
3. Agriculture
-
Decision Support Dashboard Architecture and Design for Smart Farming (Dr Wang Hui Hui)
-
Rhizomorph Mycelium Recognition using Deep Learning during Mushroom Cultivation Process - F08/PARTNERS/2128/2021 (Ts Dr Hamimah Ujir)
4. Languages and Culture
-
Comparative Analysis 3D Faces based on Expression Intensity information of Malaysian Ethnics data - Race/c(2)/1331/2016(4) (Ts Dr Hamimah Ujir)
-
3D facial features tracking for 4D expression intensity estimation - ERGS/ICT07(03)/1020/2013(17) (Ts Dr Hamimah Ujir)
- Evaluation of Existing 3D Facial Features Performance For 3D Facial Expression Using UPM-3DFE Database- SGS/02(S123)/998/2013(07) (Ts Dr Hamimah Ujir)
5. Others
-
Design of A Deep Learning Model with Attention Mechanism for Biometric Re-identification of Green Sea Turtles in Long-term Tracking Scenario, Fundamental Research Grant Scheme, 2021-09-07 to 2024-09-06 (Dr Irwandi Hipni Mohamad Hipiny)
-
Background Subtraction in Egocentric Videos Depicting Activities of Daily Living, Research Acculturation Collaborative Effort, 2015-01-26 to 2017-01-25 (Dr Irwandi Hipni Mohamad Hipiny)
-
Passive Biometric Identification of Sea Turtles (Chelonia Mydas), Small Grant Scheme, 2015-12-01 to 2016-11-30 (Dr Irwandi Hipni Mohamad Hipiny)
-
Automated Skill Determination From Egocentric Videos Containing Gaze Data, Cross Disciplinary Research, 2019-09-01 to 2021-08-31 (Dr Irwandi Hipni Mohamad Hipiny)
Member
- Professor Dr Wang Yin Chai
- Associate Professor Dr Dayang Nurfatimah Awang Iskandar, Semantic Web, Digital Image Processing System, Advancement of ICT Knowledge, Information, Computer and Communication Technology
- Associate Professor Dr Jacey-Lynn Minoi, Gameful and playful design thinking methology towards behavioural science
- Dr Chai Soo See, Remote Sensing, Predictive Analytics, Machine Learning, Graphic and Multimedia Services, Natural Hazards , Information, Computer and Communication Technology
- Dr Wang Hui Hui
- Ts Dr Hamimah Ujir, Computer Vision, Face & Gesture Recognition, Digital Competency
- Dr Irwandi Hipiny, Computer Vision, Animal Re-ID, Pattern Recognition
- Ts Dr Lim Phei Chin, Exploratory Data Analysis, Project Management, Digital Image Processing System
- Dr Fadilla 'Atyka Nor Rashid, Neural Network Computer, Machine Learning, Computer-Based Teaching and Learning
- Dr Mohammad Hossin, Decision Support System, Soft Computing, Intelligent Systems, Computer-Based Teaching and Learning
- Mr Jonathan Sidi, UiUX Design, Mobile App in Agriculture
- Miss Vanessa Wee Bui Lin
- Silvia Joseph, Iban’s Plaited Mat Motifs Recognition using Invariant Image Features
- Khalif Amir Zakry, Design of A Deep Learning Model with Attention Mechanism for Biometric re-ID of Green Sea Turtles in Long-term Tracking Scenario
- Syahiran Soria, Automatic segmentation of Region-of-Interest (ROI) in nesting Green sea turtle videos
- Ruhana Abang Yusuf, A Semantic Extraction and Analysis for Traffic Density using Traffic Images
Past Research Students
- Izzah Nilamsyukriyah Binti Buang, Malaysia Ethnicity-based Facial Expression Classification and Emotion Mapping
- Golam Morshed, Customer’s Satisfaction Facial Expression Analysis
- Wong Swee Yin, Formalize Use Case to Model the Ambiquous Scenario
- Sarawak Forestry Corporation
- Universiti Putra Malaysia (UPM), Selangor
- ARRA Mushroom Sdn. Bhd.
Dataset Link : CDRG_UNIMAS_TIKTOK | Kaggle
The dataset contains 240 videos by 20 subjects. Subjects performed 12 Tik Tok dance challenges each and captured the performance on video using their own personal camera device. Videos are ranked per label category, as voted by 100 human annotators in a pairwise ranking exercise. This dataset was produced as part of our UNIMAS Cross-disciplinary Research Grant's (2019) outputs.
Please cite the following article if you use this dataset in your publication(s):
Hipiny, I., Ujir, H., Alias, A.A., and Shanat, M. (2023). Who Danced Better? Ranked Tik Tok Dance Video Dataset And Pairwise Action Quality Assessment. Manuscript submitted for publication.
or
Zakry, K.A., Hipiny, I., and Ujir, H. (2023). Classification of dances using AlexNet, ResNet18 and SqueezeNet1_0. International Journal of Artificial Intelligence, IAES, 12(2).
Dataset Link : (PDF) PANDANCHELOMY.zip (researchgate.net)
The dataset contains:
1) Folder [images] contains raw images captured at Pandan Beach, Lundu.
2) Folder [rotated_images] contains rotated raw images so they appear upright.
3) The [rotated_images\ROIs.txt] file contains entries in the following format: Each entry contains the following two coordinates: top-leftmost (x_min, y_min) and bottom-rightmost (x_max, y_max). These two coordinates define the Region-of-Interest (ROI) inside the current image.
Please cite this paper: Hipiny, I., Ujir, H., Mujahid, A., and Yahya, N.K. (2019). Towards Automated Biometric Identification of Sea Turtles (Chelonia mydas). Journal of ICT Research and Applications 12(3), ISSN: 2337-5787, E-ISSN: 2338-5499.
