Thanh-Toan Do
Research Fellow
Australian Centre for Robotic Vision (ACRV), The University of Adelaide, Australia.
Email: thanh-toan.do(at)adelaide.edu.au

Google Scholar

  • Research interests:
    • Computer vision: robotic vision, image-based localization, large scale visual search.
    • Machine learning: deep learning, sparse coding, learning to hash.

  • 09/2018. Our paper "Non-smooth M-Estimator for Maximum Consensus Estimation" won the Best Science Paper Award at BMVC 2018. Congrats to Huu Le and the team! 
  • 08/2018. Our paper "Improving Chamfer Template Matching Using Image Segmentation" is accepted to IEEE Signal Processing Letters (IEEE SPL).
  • 07/2018. Ours papers have been accepted at BMVC 2018 (2 oral papers) and ECCV 2018 (1 paper). 
  • 05/2018. Paper on Deep Supervised Hashing is accepted at IEEE International Conference on Image Processing (ICIP), 2018
  •  02/2018: The paper on 'Melanoma Detection using Smartphones and Mobile Image Analysis' is accepted to IEEE Trans. on Multimedia (TMM).
  • I will give a tutorial on Visual Learning at "Robotic Vision Summer School" which takes place from 4th February 2018 to 9th February 2018, in Kioloa, Australia.
  • 04/2017: TPAMI paper on image search:
    • Thanh-Toan Do and Ngai-Man Cheung, "Embedding based on function approximation for large scale image search." IEEE Transactions on Pattern Analysis and Machine Intelligence 2017. [PDF] Code and precomputed features (3.5Gb) 
  • 03/2017: Hashing paper at CVPR 2017
    • Thanh-Toan Do, Dang-Khoa Le Tan, Trung Pham, Ngai-Man Cheung, "Simultaneous Feature Aggregating and Hashing for Large-scale Image Search," in Proc. IEEE CVPR 2017. [PDF].