BLOG
31 March 2021
Visual localization is used by robots and self-driving cars to estimate their position and in AR apps to interact with the physical world both indoors and out in the open. This article gives an overview of current state-of-the-art methods and their advantages and drawbacks.
2021
8 December 2020
How harder negatives facilitate better and faster contrastive self-supervised learning and ways of synthesizing harder negative features on-the-fly.
2020
24 August 2020
How mid-size sets of captioned images can rival with large-scale labelled image sets to learn generic representations
2020
12 June 2020
A first-of-its-kind architecture that, based on a single image, predicts how a robot can pick up objects from within any scene could revolutionize applications in AR/VR and robotics.
2020
29 January 2020
Latest release of the popular synthetic image dataset for training and testing. New features include increased photorealism, stereo cameras and additional ground truth.
2020
4 December 2019
Combining keypoint reliability in an image as part of the keypoint detection problem significantly improves feature matching.
2019
25 October 2019
Evaluating our method on public datasets, we show that it can successfully solve challenging situations in dynamic environments which cause state-of-the-art baseline VSLAM algorithms to fail and that it maintains performance on static scenes.
2019
18 October 2019
Fine-grained action retrieval made possible thanks to a new annotated dataset and part-of-speech embedding.
2019
14 June 2019
The acme of perfection of AR is when you can’t distinguish the virtual from the real. We experiment with AI and computer vision in our AR museum guide ARAO to make the physical experience a natural one.
2019
7 June 2019
This blog presents our CVPR’19 paper on “MARS: Motion-Augmented RGB Stream for Action Recognition” done with the Thoth team at Inria.
2019
27 May 2019
The first deep regression-based method that generalizes to larger environments, performs better than state-of-the-art with little training data and is robust to occlusion. Virtual Gallery dataset provided.
2019
28 February 2019
An article we recently published on arXiv gives an overview of the evolution of local features - from handcrafted to deep-learning-based methods - and a discussion of several benchmarks and papers that evaluate local features.
2019