Additional Information Page for ICASSP 2019 Submission Under Review

This webpage contains additional information for our ICASSP 2019 submitted paper, titled, LEARNING SOUND EVENTS FROM WEBLY LABELED DATA .

Authors: Anurag Kumar, Ankit Shah, Alex Hauptmann, Bhiksha Raj

Email: alnu AT andrew DOT cmu DOT edu, aps1 AT andrew DOT cmu DOT edu

If you did not come on this page from the paper, might be a good idea to read the paper first, pdf

We provide addtional information here.

This paper is about learning from web data without involving human supervision in the process.


Code

Coming Soon.

Webly Labeled Data

1. Will be available for download soon.

2. Click Here For More Details Webly-2k and Webly-4k datasets.


System

WeblyNet
WeblyNet system to learn from webley labeled data


WeblyNet


Some Additional Results and Analysis

Class specific results and some additional discussion is available below.

Click Here For More Details and Results


References

[1] Anurag Kumar, Bhiksha Raj , “Audio Event Detection using Weakly Labeled Data,” ACM Multimedia (MM), 2016
[2] Anurag Kumar, Bhiksha Raj , “Audio Event and Scene Recognition: A Unified Approach using Strongly and Weakly Labeled Data ” IEEE IJCNN, 2017