I'm presently working as a Scientist in the Institute for Infocomm Research (I2R) division of A*STAR, Singapore. I obtained my Ph.D. at the National Univeristy of Singapore, under the joint supervision of Prof. Robby T. Tan and Prof. Loong-Fah Cheong. I've also worked in collaboration with Dr. Lionel Heng from DSO National Labortories, Singapore. My primary research interests include Computer Vision and Deep Learning.

Previously, I worked as a senior design engineer at Freescale Semiconductors, India. I obtained my B.Eng. from Delhi College of Engineering, University of Delhi, India.


My previous research works were dedicated to: (1) Handling low-level vision problems such as depth from stereo and optical flow estimation, and (2) Performing visibility enhancement, under degraded visibility conditions such as nighttime and daytime fog. Related publications are presented below.

DC-ShadowNet: Single-Image Hard and Soft Shadow Removal Using Unsupervised Domain-Classifier Guided Network
Yeying Jin, Aashish Sharma, Robby T. Tan
International Conference on Computer Vision (ICCV), 2021, Montreal, Canada
paper | bibtex | code

Nighttime Visibility Enhancement by Increasing the Dynamic Range and Suppression of Light Effects
Aashish Sharma, Robby T. Tan
Computer Vision and Pattern Recognition (CVPR), 2021, Nashville, USA
paper | bibtex

Nighttime Stereo Depth Estimation using Joint Translation-Stereo Learning: Light Effects and Uninformative Regions [Oral]
Aashish Sharma, Loong-Fah Cheong, Lionel Heng, Robby T. Tan
International Conference on 3D Vision (3DV), 2020, Fukuoka, Japan
paper | bibtex | arXiv'19 | code (arXiv'19)

Single-Image Camera Response Function Using Prediction Consistency and Gradual Refinement
Aashish Sharma, Robby T. Tan, Loong-Fah Cheong
Asian Conference on Computer Vision (ACCV), 2020, Kyoto, Japan
paper | bibtex

Optical Flow in Dense Foggy Scenes using Semi-Supervised Learning
Wending Yan*, Aashish Sharma*, Robby T. Tan (*equal contribution)
Computer Vision and Pattern Recognition (CVPR), 2020, Seattle, USA
paper | bibtex

Optical flow estimation under dense fog conditions via domain adaptive semi-supervised learning using labelled synthetic and unlabelled real fog data.

Into the Twilight Zone: Depth Estimation using Joint Structure-Stereo Optimization
Aashish Sharma, Loong-Fah Cheong
European Conference on Computer Vision (ECCV), 2018, Munich, Germany
paper | poster | bibtex

Depth estimation for weakly-lit nighttime images by optimizing a joint structure-structure model.

Academic Services

Reviewer: CVPR-V4AS’19, PSIVT’19, IJCV’20, MVA’21, 3DV’21


  • Feb’2017-May’2021: Ph.D. (Computer Vision and Deep Learning), NUS, Singapore
  • Aug’2008-Jun’2012: B.Eng.(Electronics and Communication), DCE, Delhi University, India

Work Experience

  • Jun’2021- : Scientist at I2R, A*STAR, Singapore
  • Feb’2017-May’2021: Research Engineer at NUS, Singapore
  • Jun’2012-May’2016: Senior Design Engineer at Freescale Semiconductors, India