person · 0.982
face · 0.961
REC ●track_id 07DEPTH1.37m
I'm a computer vision researcher from Bangladesh, based in South Korea. My work spans traditional and deep-learning methods for 2D and 3D image processing, generative modeling and multi-sensor fusion. These days I focus on driver monitoring and in-cabin perception for production automotive systems.
Monocular, stereo and event based depth, 3D reconstruction and sensor fusion across RGB, IR, LiDAR, RADAR and IMU.
Image inpainting, HDR reconstruction, super resolution and reversed ISP restoration with attention and generative models.
DMS and OMS pipelines: gaze, 3D body keypoints, seat pose and child presence detection for vehicle safety.
Pedestrian and obstacle detection, distance and trajectory estimation, real time perception on edge devices.
Monocular depth estimation for in-cabin monitoring units.
Pedestrian detection, distance estimation and predictive trajectory estimation for a Level-4 PoC ADAS system.
Real-time eye distance estimation from a glasses-free 3D light-field display.
Detection, distance estimation, localization and tracking of workers using a long-wave infrared camera.
Deep stereo matching and motion deblurring algorithms using a stereo event camera.
High-speed machine-learning detection engine for real-time pupil tracking of a holographic display.
Open to research collaborations and hard perception problems. The fastest way to reach me is email.
Say hello →