“Understanding the world using visual data.”
Computer vision is the discipline concerned with making the best use of visual data and understanding the world through signals, images and videos. It provides enabling technologies for innovative applications and services in a wide spectrum of industry sectors including automotive, agribusiness, defense and aerospace, environment, information and telecommunication, manufacturing, medical devices, surveillance, and transportation.
Our goal is to excel in research and then bridge it to real-world applications of commercial value for wealth creation in Australia. Our ambition is to “be the number one in what we do” globally.
To achieve this, we contemplate:
• building upon our core research strengths and growing around them,
• searching for creative ideas to improve our research base for new projects,
• increasing research capacity in strategic directions to respond industry and government needs in a dynamic way,
• involving in all stages of innovative process to acquire insight and actionable information,
• sharing knowledge to advance our research strategies,
• consulting each other in decision-making,
• acting swiftly on actionable information when we see the opportunity,
• documenting our ideas and progress to help clarify thinking and sustainable development.
CVRG values these principles:
• raise the stars in computer vision discipline
• believe 1+1 = 11: talent wins games, but teamwork wins championships!
• none of us is as smart as all of us: do things in consultation.
• nurture open and fluid communication.
• show respect to the opinions of others.
• not how many hours being worked but what is generated.
Prof Fatih Porikli
Leader, Computer Vision Research Group
- Parametric Learning of Texture Filters by Stacked Fisher Autoencoders
- Less is More: Towards Compact CNNs
- Ultra Resolution by Discriminative Generative Networks
- Building Scene Models by Completing and Hallucinating Depth and Semantics
- Tensor Representations via Kernel Linearization for Action Recognition from 3D Skeletons