Major Research Area

Computational Photography & Imaging 

• • Image restoration, generative models for image synthesis 

  • Multi-camera imaging systems, emerging image sensors 

3D Vision & Depth Perception for Spatial Intelligence in Machines

Neuromorphic Event-based Vision 

• • Supervised and unsupervised deep learning models for event camera 

Deep Learning for Visual AI 

• • Supervised and unsupervised deep learning models for image understanding

Research Topics

Image Restoration 

• • Image restoration (motion deblurring, inpainting, colorization, denoising, reflection removal, dehazing, camera ISP, etc.)

  • Image fusion (low-light enhancement, HDR imaging), multi-modal fusion (text-guided image editing)

  • Based on generative models for image synthesis such as GAN and diffusion models 

Event-based Vision 

• • Event camera-based depth estimation, event-based image/video deblurring

  • Scene recognition using event cameras (event-based object detection)

Depth Estimation 

• • Depth from single (monocular) view, Depth from stereo image pair, Structure from motion

  • Depth cue fusion and depth sensor fusion

View Synthesis  

• • Multi-view synthesis from stereo image pairs

  • Synthesis from multi-view images using Nerf and 3D Gaussian Splatting (GS)

Segmentation & Object Detection

• • Supervised and unsupervised deep learning models for object recognition

  • Multi-modal feature fusion (event-image)

    Defense AI for military applications

• • Intelligent radio system, spectrum sensing (such as automatic modulation classification)

  • Object recognition for unmanned aerial vehicle