ECCV 2020 Tutorial on Weakly-Supervised Learning in Computer Vision
Deep convolutional networks have become the go-to technique for a variety of computer vision task such as image classification, object detection, segmentation, key points detection, etc. These over-parameterized models are known to be data-hungry; tens of thousand of labelled examples are typically required. Since manual annotations are expensive, learning from “weaker” annotations (e.g. only image-level category labels to localize object instances by a bounding box) become key to expand the success of deep networks to new applications.
This tutorial will provide an overview of weakly supervised learning methods in computer vision, and we will discuss the broad area of weakly supervised object recognition and its limitations of current state-of-the-art, evaluation metrics, and future ideas that will spur disruptive progress in the field of weakly supervised learning.
|Hakan Bilen||Rodrigo Benenson||Seong Joon Oh|
|1-Introduction||Hakan Bilen||15 min||ECCV2020 / Youtube / Slides(2.5MB)|
|2-Weakly Supervised Object Detection||Hakan Bilen||60 min||ECCV2020 / Youtube / Slides(3.4MB)|
|3-Weakly Supervised Semantic Segmentation||Rodrigo Benenson||45 min||ECCV2020 / Youtube / Slides(15MB)|
|4-Human-in-the-Loop Annotations||Rodrigo Benenson||25 min||ECCV2020 / Youtube / Slides(3.8MB)|
|5-Evaluating Weakly-Supervised Methods||Seong Joon Oh||35 min||ECCV2020 / Youtube / Slides(3.0MB)|
|6-Closing Remarks||Seong Joon Oh||10 min||ECCV2020 / Youtube / Slides(1.7MB)|
Please use the form below to ask any question related to the topic - we will try our best to answer all the questions during the Q&A slot.
You can ask either very specific or very open-ended questions. You can send multiple questions. You may as well choose to be anonymous in the form. In short, don’t be shy to ask questions!
During the live sessions we will dedicate time to answer the questions asked via this form, and we will also take questions live.