2021 Seminars


  • 26 July 2021 (2.00pm, Monday, Online) Ms Bingjie Xiao (CS, AUT) Topics on “Fruit Recognition from Digital Images by Using Deep Learning
  • Automatic classification of fruits by using computer vision and deep learning is one of the pivotal steps towards fruit separation and cleaning in a production line. The precise positioning and accurate recognition of fruits will lead to high efficiency in NZ fruit industry and ramp up local economy. In this proposal, we conduct fruit detection and classification, deep learning models are able to assist us in fruit classification, which allow us to utilise digital images from networked cameras to classify fruits automatically. The goal of this project is to verify the state-of-the-art deep learning models for fruit classification so as to lessen the onerous work and reduce human labour costs. Our methodology consists of four parts, namely, data collection, digital image processing, fruit detection and classification, ripeness identification, and outcome evaluations. So far, we have collected and annotated images as our dataset. Pertaining to implement fruit detection and classification from digital videos, we have experimented with the one-stage and two-stage methods in deep learning for visual object detection, we have taken use of a plethora of our proposed methods to improve the newly emerging models such as YOLOv5, etc. At present, the contribution of our project is to find that deep learning classifiers are able to achieve the best result, e.g., the ripeness of an apple from a given digital image is able to be precisely identified. We will optimise our deep learning models so as to achieve the best outcome for fruit ripeness identification in future. 
  • 20 May 2021 – 29 July 2021(Thursday, Online) Topics on “Deep Learning and Its Applications”
  • (1) Vehicle-Related Scene Understanding (X. Liu, 20 May 2021, 12pm)
  • (2) Privacy Preservation Using Deep Learning (B. Ma, 27 May 2021, 12pm)
  • (3) 3D Object Detection Using Deep Learning (S. Mahtab, 3 June 2021, 12pm)
  • (4) Fruit Detection from Digital Images (B. Xiao, 10 June 2021, 12pm)
  • (5) Visual Blockchain for Smart City (K. Gadara, 17 June 2021, 12pm)
  • (6) Cancer Recognition (F. Younus, 24 June 2021, 12pm)
  • (7) Digital Image Color Transfer (Y. Wang, 1 July 2021, 12pm)
  • (8) Automatic Speech Recognition Using Deep Learning (S. Liang, 1 July 2021, 12pm)
  • (9) Traffic sign recognition (J. Xing, 8 July 2021, 2pm)
  • (10) Braille Recognition (C. Li, 8 July 2021, 2pm)
  • (11) Swimmer Identification (X. Cao, 15 July 2021, 12pm)
  • (12) Surveillance Event Forecasting: Crime Prediction (J. Liu, 15 July 2021, 12pm)
  • (13) Currency Transparent Logo Recognition (D. Tong, 22 July 2021, 12pm)
  • (14) Image Captioning (M. Sheng, 22 July 2021,12pm)
  • (15) Traffic Sign Recognition Based on Deep Learning (Y. Zhu, 29 July 2021)
  • (16) Facial Image Super-resolution (L. Tang, 29 July 2021)
  • 23 April 2021-28 May 2021 (6.00pm, Friday, Online) Wei Qi Yan (CS, AUT) Topics on “Deep Learning and Its Development Platforms”
  • (1) Deep Learning and Data Visualisation (23 April 2021)
  • (2) Reinforcement Learning and Software WEKA (07 May 2021)
  • (3) Generative Adversarial Nets (GANs) and Software R (14 May 2021)
  • (4) AutoEncoder and Software NeuCube (21 May 2021)
  • (5) Manifold Learning and TensorFlow/TensorBoard (28 May 2021)
  • 25 March 2021 (1.00pm, Thursday, Online) Dr Jia Lu (CS, AUT) Topics on “Deep Learning Methods for Human Behaviour Recognition”
  • Video surveillance has been broadly applied to public places which allows security staff to monitor abnormal events. However, most of the video surveillance software platforms are still being run in traditional mode. As the state-of-the-art technology, deep learning (DL) becomes much popular because of its superiority to conventional machine learning. Moreover, deep learning as an end-to-end model normally does not require low-level processing which is able to cut off human labour and gain time efficiency for human behaviour recognition though the training is costly. In this project, we design and implement deep learning methods which are time efficiency and outperform in training and testing. As the outcome of this research work, we attain an overall up to 90% accuracy of recognition in real-time. Moreover, we have created our own dataset for the experiments. 
  • 23 March 2021 (1.00pm, Tuesday, Online) Ms Xuejing Chen (CS, AUT) Topics on “Deep Learning Approach to Spatio-Temporal Image Recognition for Road Surface Damage Detection”
  • Road damages like cracks, alligators, and potholes are critical problems for road constructions and major considerations for driving decisions. Road pavement distresses, at the early stage of road damage formation, also signal needs for road maintenance. Recently computer-vision techniques for detecting such damages mostly rely on convolutional neural networks which have been studied but not reached good results, yet still facing the limitations related to localisation, 3D consideration, and run-time efficiency. In this research project, we propose a deep learning architecture to efficiently learn, detect, and reconstruct 3D road damages concerning different quality and temporal-spatial information of streaming images. We have conducted an experimental study and the results show that our initially proposed architecture of Ensembled Aggregated Pyramid Residual Network achieves better performance than the state-of-the-art CNN-based models. Based on this, we continue improving and evaluating our models based on BDD and CCSAD to extract temporal spatial information as well as 3D images of road damage objects. We will also perform temporal-spatial causality analysis and stereo measures to obtain a real-time and fine-grained road surface damage measurement. Our research work will advance the field and can help transportation agencies make informed decisions on scheduling a repair or rebuilding at an appropriate time.

  • 11 Feb 2021 -22 April 2021 (12.00pm, Thursday, Online) Wei Qi Yan (CS, AUT) Topics on “Postgraduate Mathematics and Reinforcement Learning”
  • (1) Functional Analysis: Part 1 (11 February 2021)
  • (2) Functional Analysis: Part 2 (18 February 2021)
  • (3) Finite Fields (25 February 2021)
  • (4) Manifold Learning (04 March 2021)
  • (5) Probabilistic Graphical Models (11 March 2021)
  • (6) Generative Adversarial Networks (18 March 2021)
  • (7) Autoencoder (25 March 2021)
  • (8) Reinforcement Learning (01 April 2021)
  • (9) RL1-1: Exact Solution Methods (08 April 2021)
  • (10) RL1-2: Exact Solution Methods (15 April 2021)
  • (11) RL2-1: Approximate Solution Methods (22 April 2021)
  • (11) RL2-2: Approximate Solution Methods (06 May 2021)
  • (12) RL3: Looking Deeper Methods (13 May 2021)