Football Stats Quiz
About a month ago I watched a very awesome and inspirational talk by Patrick Lucey from StatsPerform on an overview of AI in sport. In one of their slides they mention that āspo...
Interview for HydraNets course
Last month the creator of the HydraNets course on multi-task learning Jeremy Cohen reached out to me and asked if I would be interested in speaking with him about how our own mu...
Example of speeding up inference of PyTorch models on M1 via Core ML tools
I recently read the CVPR 2022 paper titled āLearning to generate line drawings that convey geometry and semanticsā, and I found the results quite interesting. Thankfully, the au...
Number of Citations of Papers Awarded at CVPR conferences
CVPR has long been regarded as the cream of the crop in the area of computer vision, attracting high-quality and influential publications. Traditionally, several works have been...
My internship at Skydio
Recently I returned to Australia from the US where I had spent five months as an intern at Skydio, a Redwood City-based startup that is building autonomous flying robots, aka dr...
Are we there yet? Part I. Education
As of 2019, if you were to build a text generator (akin to this one), chances are that after the word Artificial it would mostly likely be predicting the word Intelligence; that...
A Couple of Updates
I am currently in Long Beach attending CVPR 2019, where we are going to present our paper on fast neural architecture search for semantic segmentation ā transferrable to othe...
Tutorial - Converting a PyTorch model to TensorFlow.js
In this tutorial, I will cover one possible way of converting a PyTorch model into TensorFlow.js. This conversion will allow us to embed our model into a web-page. Someone might...
Multi-Task RefineNet Demo in Your Browser
I have just added a demo of Multi-Task RefineNet running in your browser with the help of TensorFlow.js. Check it out in demos! For now, only browsers with WebGL are supported.
Release of DenseTorch
I have just released a PyTorch wrapper that aims to facilitate a typical training workflow of dense per-pixel tasks. The project code is available here. Currently, two training ...
Real-Time Joint Segmentation, Depth and Surface Normals Estimation
Our paper, titled āReal-Time Joint Semantic Segmentation and Depth Estimation Using Asymmetric Annotationsā has recently been accepted at International Conference on Robotics an...
On the perils of being slow in semantic segmentation
Segmenting a turtle - one of the slowest animals on Earth: left - original photo, right - segmentation result (highlighted in different col...
Light-Weight RefineNet
After nearly two years from my first publication (at the same venue as now), which included a year of academic break, I have finally submitted my first paper in my new PhD journ...