Scene Understanding

In this project, we intend to tackle the problem of Scene Understanding. In other words, given the image of a scene, the objective is to extract highly semantic information, for instance, the class it belongs to (e.g., bathroom, church, library). A scene consists in places where humans can act within or navigate. Therefore the capacity to sense the environment in a more meaningful way is highly relevant for most applications that intend to operate in the real world.

Publications:

A Robust Indoor Scene Recognition Method based on Sparse Representation (CIARP 2017)

22nd Ibero-American Congress on Pattern Recognition

Abstract

In this paper, we present a robust method for scene recognition, which leverages Convolutional Neural Networks (CNNs) features and Sparse Coding setting by creating a new representation of indoor scenes. Although CNNs highly benefited the fields of computer vision and pattern recognition, convolutional layers adjust weights on a global-approach, which might lead to losing important local details such as objects and small structures. Our proposed scene representation relies on both: global features that mostly refers to environment’s structure, and local features that are sparsely combined to capture characteristics of common objects of a given scene. This new representation is based on fragments of the scene and leverages features extracted by CNNs. The experimental evaluation shows that the resulting representation outperforms previous scene recognition methods on Scene15 and MIT67 datasets, and performs competitively on SUN397, while being highly robust to perturbations in the input image such as noise and occlusion.

Methodology Diagram

Links

            Source code – Available here!

            Paper Draft – Available here!

Citation

@inproceedings{Nascimento2017,
Title = {A Robust Indoor Scene Recognition Method based on Sparse Representation},
Author = {Nascimento, Guilherme and Laranjeira, Camila and Braz, Vinicius and Lacerda, Anisio and Nascimento, Erickson Rangel},
booktitle = {22nd Iberoamerican Congress on Pattern Recognition. CIARP},
Publisher = {Springer International Publishing},
Year = {2017},
Address = {Valparaiso, CL},
note = {To appear},
}

Presentation Video

Coming soon!

Team: