Active nerve cells

Accurately counting maize tassels is important for monitoring the growth status of maize plants. This tedious task, however, is still mainly done by manual efforts. In the context of modern plant phenotyping, automating this task is required to meet the need of large-scale analysis of genotype and phenotype. In recent years, computer vision technologies have experienced a significant breakthrough due to the emergence of large-scale datasets and increased computational resources. Naturally image-based approaches have also received much attention in plant-related studies. Yet a fact is that most image-based systems for plant phenotyping are deployed under controlled laboratory environment. When transferring the application scenario to unconstrained in-field conditions, intrinsic and extrinsic variations in the wild pose great challenges for accurate counting of maize tassels, which goes beyond the ability of conventional image processing techniques. This calls for further robust computer vision approaches to address in-field variations. This paper studies the in-field counting problem of maize tassels. To our knowledge, this is the first time that a plant-related counting problem is considered using computer vision technologies under unconstrained field-based environment.

This results is published in the Plant Methods journal.
PDF is here

Screenshot 2017-10-07 20.49.24

counting

Posted in deep learning, publications, research
Leave a comment

This work was presented in IEEE Conf Computer Vision Pattern Recognition 2017 in Hawaii

Posted in deep learning, research
Leave a comment

The slides can be accessed here

Posted in news, research
Leave a comment

Guosheng will be an Assistant Professor at Nanyang Tech. Uni, Singapore; and Yao will be joining the Amazon Go team at Seattle.

Posted in news
Leave a comment

Adelaide team is No. 1 on Cityscapes semantic segmentation. See our paper. The Cityscapes dataset is a large-scale dataset for semantic urban scene understanding, containing a diverse set of street scene video recordings from 50 cities. It’s one of the important benchmark datasets for autonomous driving, developed by Daimler AG.

Posted in deep learning, publications, research
Leave a comment

The Adelaide Machine Learning Group has developed a unified image processing technique based on Deep Learning, which can be applied to various image enhancement tasks such as, not limited to, Image Denoising, Image Inpainting, Image Super-resolution, Image Deblur. This work will be presented at The Thirtieth Annual Conference on Neural Information Processing Systems (NIPS), Spain, […]

Posted in deep learning, publications, research
Leave a comment

Adelaide team attended ImageNet Challenge 2016 on the task of Scene Parsing and won the 2nd place. The ImageNet Scene Parsing competition requires entrants to correctly label each pixel into 150 categories (building, road, car, computer, toy shop, cat, horse, and so on) for photographs from Flickr and various search engines with as few errors […]

Posted in deep learning, research
Leave a comment

As of March 2016, we achieved No. 1 on ICDAR 2015 Challenge 2 “Focused Scene Text” for the task of End-to-End Scene Text Recognition. See the Leaderboard. Details can be found in the technical report.

Posted in deep learning, publications, research | Tagged ,
Leave a comment

We design new methods for semantic pixel labelling, which set the new record. Details can be found in our technical report. See the PASCAL VOC Leaderboard.

Posted in deep learning, publications, research | Tagged ,
Leave a comment