THE LAB, RESEARCH & DEVELOPMENT
At Heuritech, we specialize in Deep Learning and Computer Vision. We have created this lab, dedicated to Artificial Intelligence and Sustainability.
The Lab: Artificial Intelligence for Sustainability
At Heuritech, we believe that climate change and environmental issues should be a priority for all companies. While our main focus is helping fashion and beauty brands to produce more mindfully and reduce their overstock, we all want to have a more direct impact, through The Lab.
Heuritech is proud to announce its commitment to helping NGO, Surfrider Foundation Europe, reduce plastic ocean pollution. Created in 1990, Surfrider is one of the leading ocean protection foundations. The partnership consists of using Heuritech’s cutting-edge image recognition technology to detect plastic pollution in pictures of rivers taken by Surfrider members for pollution analysis purposes.
A unique R&D approach
Research is at the heart of Heuritech, as its 2 founders hold a PhD in Machine Learning.
Our R&D team is an interesting combination of research and engineering. The research aspect consists of staying at the forefront of Deep Learning and Computer Vision by publishing papers at top conferences (CVPR, ICCV, …), and organizing conferences in Paris. Additionally, we work toward viable engineering solutions where we aim to use our latest developments to build a Computer Vision pipeline that can analyze all clothing very precisely, at scale, from millions of images each day. Since typical Deep Learning open sources libraries do not work for unconstrained images, handling the unexpected is the main topic of our research. Read about our meetups, our lectures and our published research papers in this blog.
OUR RESEARCH PAPERS ON IMAGE ANALYSIS & DEEP LEARNING
PLOP: Learning without Forgetting for Continual Semantic Segmentation
– CVPR 2021 –
Insights from the Future for Continual Learning
– CVPR CLVISION Workshop 2021 –
Continuum: Simple Management of Complex Continual Learning Scenarios
PODNet: Small-Tasks Incremental Learning
MUREL: Multimodal Relational Reasoning for Visual Question Answering
– CVPR 2019 –
BLOCK: Bilinear Superdiagonal Fusion for Visual Question Answering and Visual Relationship Detection
– AAA 2019 –
OMNIA Faster R-CNN: Detection in the wild through dataset merging and soft distillation
Leveraging Weakly Annotated Data for Fashion Image Retrieval and Label Prediction
– ICCV Fashion Workshop 2017 –
MUTAN: Multimodal Tucker Fusion for Visual Question Answering
– ICCV 2017 –
DEEP LEARNING MEETUPS
Heuritech is the founder and organizer of Deep-Learning Paris Meetup, a community of 5000 researchers gathering once a month.