Hasan Firdaus Mohd Zaki

About

Hasan Firdaus Mohd Zaki is a researcher at the forefront of computer vision and machine learning, with a primary focus on RGB-D perception and environmental sound classification. His major contributions lie in developing efficient, viewpoint-invariant methods for semantic scene and object category recognition using multimodal RGB-D data. Notably, his work on "Learning a deeply supervised multi-modal RGB-D embedding" (2017, 20 citations) advanced the integration of depth and color information for robust recognition. He further pioneered optimized parameter tuning for convolutional neural networks in sound classification (2021, 19 citations), bridging vision and audio domains. Zaki's research on "Localized Deep Extreme Learning Machines" (2015, 9 citations) addressed critical challenges in real-time robotics by reducing training time and data requirements for RGB-D object recognition, offering a practical alternative to traditional deep networks. His viewpoint-invariant approach (2018, 16 citations) has been instrumental in making object and scene categorization more robust in dynamic environments. With a cumulative impact of over 64 citations across his key works, Zaki's contributions are shaping more efficient, real-world deployable AI systems for robotics and autonomous perception.

Research Focus

Key Achievements

Top Papers

1. 1
   Learning a deeply supervised multi-modal RGB-D embedding for semantic scene and object category recognition

   20 citations · 2017
2. 2
   Rethinking environmental sound classification using convolutional neural networks: optimized parameter tuning of single feature extraction

   19 citations · 2021
3. 3
   Viewpoint invariant semantic object and scene categorization with RGB-D sensors

   16 citations · 2018
4. 4
   Localized Deep Extreme Learning Machines for Efficient RGB-D Object Recognition

   9 citations · 2015

Key Collaborators

Contact & Links