- Link: https://github.com/lannguyen0910/food-recognition
- Summary:
We deployed a web application where user can upload images or videos of their dishes to get full analysis of the ingredients and nutritions of it. Food ingredients are detected using object detection deep learning algorithm and their information is retrieved from online APIs.
- Tags: detection, classification, web-application
- Link: https://github.com/kaylode/shrec22-pothole
- Summary:
We implemented semi-supervised algorithms to detect and segment road cracks and street potholes in the wild
- Tags: 2D object segmentation, cross pseudo supervision
- Link: (private)
- Summary:
We deployed a system to automatically detection and classify traffic signboards in panoramic Vietnam-street images. This work is a collaboration between MapDAS and HCMUS
- Tags: detection, classification
Semi-supervised organ segmentation with Mask Propagation Refinement and Uncertainty Estimation for Data Generation
- Link: https://github.com/kaylode/ivos
- Summary:
We present a novel two-staged method that employs various 2D-based techniques to deal with the 3D segmentation task. In most of the previous challenges, it is unlikely for 2D CNNs to be comparable with other 3D CNNs since 2D models can hardly capture temporal information. In light of that, we propose using the recent state-of-the-art technique in video object segmentation, combining it with other semi-supervised training techniques to leverage the extensive unlabeled data. Moreover, we introduce a way to generate pseudo-labeled data that is both plausible and consistent for further retraining by using uncertainty estimation.
- Tags: 2D object segmentation, cross pseudo supervision, label propagation
