From Photo Streams to Evolving Situations
This work addresses the challenge of detecting and clustering situations from noisy, multi-source photo data, representing an incremental advancement in semi-supervised learning for visual recognition.
The paper tackles the problem of evolving situation recognition from photo streams by developing a semi-supervised learning framework with graph-based models and noise-robust norms, achieving improved accuracy on the Yahoo Flickr Creative Commons 100 Million dataset.
Photos are becoming spontaneous, objective, and universal sources of information. This paper develops evolving situation recognition using photo streams coming from disparate sources combined with the advances of deep learning. Using visual concepts in photos together with space and time information, we formulate the situation detection into a semi-supervised learning framework and propose new graph-based models to solve the problem. To extend the method for unknown situations, we introduce a soft label method which enables the traditional semi-supervised learning framework to accurately predict predefined labels as well as effectively form new clusters. To overcome the noisy data which degrades graph quality, leading to poor recognition results, we take advantage of two kinds of noise-robust norms which can eliminate the adverse effects of outliers in visual concepts and improve the accuracy of situation recognition. Finally, we demonstrate the idea and the effectiveness of the proposed model on Yahoo Flickr Creative Commons 100 Million.