4 results on '"Kosti, Ronak"'
Search Results
2. Visual scene context in emotion perception
- Author
-
Kosti, Ronak, Lapedriza García, Àgata, Lapedriza Garcia, Àgata, and Universitat Oberta de Catalunya. Escola de Doctorat
- Subjects
visió per ordinador ,reconeixement de les emocions ,emotion recognition ,visión por ordenador ,informàtica afectiva ,informática afectiva ,Computer vision ,affective computing ,reconocimiento de las emociones - Abstract
Els estudis psicològics demostren que el context de l'escena, a més de l'expressió facial i la postura corporal, aporta informació important a la nostra percepció de les emocions de les persones. Tot i això, el processament del context per al reconeixement automàtic de les emocions no s'ha explorat a fons, en part per la manca de dades adequades. En aquesta tesi presentem EMOTIC, un conjunt de dades d'imatges de persones en situacions naturals i diferents anotades amb la seva aparent emoció. La base de dades EMOTIC combina dos tipus de representació d'emocions diferents: (1) un conjunt de 26 categories d'emoció i (2) les dimensions contínues valència, excitació i dominància. També presentem una anàlisi estadística i algorítmica detallada del conjunt de dades juntament amb l'anàlisi d'acords d'anotadors. Els models CNN estan formats per EMOTIC, combinant característiques de la persona amb funcions d'escena (context). Els nostres resultats mostren com el context d'escena aporta informació important per reconèixer automàticament els estats emocionals i motiven més recerca en aquesta direcció. Los estudios psicológicos muestran que el contexto de la escena, además de la expresión facial y la pose corporal, aporta información importante a nuestra percepción de las emociones de las personas. Sin embargo, el procesamiento del contexto para el reconocimiento automático de emociones no se ha explorado en profundidad, en parte debido a la falta de datos adecuados. En esta tesis presentamos EMOTIC, un conjunto de datos de imágenes de personas en situaciones naturales y diferentes anotadas con su aparente emoción. La base de datos EMOTIC combina dos tipos diferentes de representación de emociones: (1) un conjunto de 26 categorías de emociones y (2) las dimensiones continuas de valencia, excitación y dominación. También presentamos un análisis estadístico y algorítmico detallado del conjunto de datos junto con el análisis de concordancia de los anotadores. Los modelos CNN están entrenados en EMOTIC, combinando características de la persona con características de escena (contexto). Nuestros resultados muestran cómo el contexto de la escena aporta información importante para reconocer automáticamente los estados emocionales, lo cual motiva más investigaciones en esta dirección. Psychological studies show that the context of a setting, in addition to facial expression and body language, lends important information that conditions our perception of people's emotions. However, context's processing in the case of automatic emotion recognition has not been explored in depth, partly due to the lack of sufficient data. In this thesis we present EMOTIC, a dataset of images of people in various natural scenarios annotated with their apparent emotion. The EMOTIC database combines two different types of emotion representation: (1) a set of 26 emotion categories, and (2) the continuous dimensions of valence, arousal and dominance. We also present a detailed statistical and algorithmic analysis of the dataset along with the annotators' agreement analysis. CNN models are trained using EMOTIC, combining a person's features with those of the setting (context). Our results not only show how the context of a setting contributes important information for automatically recognizing emotional states but also promote further research in this direction.
- Published
- 2019
3. Context Based Emotion Recognition Using EMOTIC Dataset.
- Author
-
Kosti, Ronak, Alvarez, Jose M., Recasens, Adria, and Lapedriza, Agata
- Subjects
- *
EMOTION recognition , *COMPUTER vision , *EMOTIONAL state , *AFFECTIVE computing , *FACIAL expression - Abstract
In our everyday lives and social interactions we often try to perceive the emotional states of people. There has been a lot of research in providing machines with a similar capacity of recognizing emotions. From a computer vision perspective, most of the previous efforts have been focusing in analyzing the facial expressions and, in some cases, also the body pose. Some of these methods work remarkably well in specific settings. However, their performance is limited in natural, unconstrained environments. Psychological studies show that the scene context, in addition to facial expression and body pose, provides important information to our perception of people's emotions. However, the processing of the context for automatic emotion recognition has not been explored in depth, partly due to the lack of proper data. In this paper we present EMOTIC, a dataset of images of people in a diverse set of natural situations, annotated with their apparent emotion. The EMOTIC dataset combines two different types of emotion representation: (1) a set of 26 discrete categories, and (2) the continuous dimensions Valence, Arousal, and Dominance. We also present a detailed statistical and algorithmic analysis of the dataset along with annotators’ agreement analysis. Using the EMOTIC dataset we train different CNN models for emotion recognition, combining the information of the bounding box containing the person with the contextual information extracted from the scene. Our results show how scene context provides important information to automatically recognize emotional states and motivate further research in this direction. [ABSTRACT FROM AUTHOR]
- Published
- 2020
- Full Text
- View/download PDF
4. ICC++: Explainable feature learning for art history using image compositions.
- Author
-
Madhu, Prathmesh, Marquart, Tilman, Kosti, Ronak, Suckow, Dirk, Bell, Peter, Maier, Andreas, and Christlein, Vincent
- Subjects
- *
ART history , *CONTENT-based image retrieval , *DIGITAL humanities , *ARTISTIC style , *DIGITAL learning , *DEEP learning , *DIGITIZATION - Abstract
• ICC++ is explainable and content-based image retrieval method. • ICC++ is based on compositional patterns for semantic understanding of art historical images. • Introducing Web Gallery of Art 500 (WGA500) dataset - to encourage further research in this domain. • Rigorous evaluations against traditional methods, deep features & SOTA methods. • Our method outperforms SOTA, while has an advantage over deep features in terms of explainability. Image compositions are helpful in the study of image structures and assist in discovering the semantics of the underlying scene portrayed across art forms and styles. With the digitization of artworks in recent years, thousands of images of a particular scene or narrative could potentially be linked together. However, manually linking this data with consistent objectiveness can be a highly challenging and time-consuming task. In this work, we present a novel approach called Image Composition Canvas (ICC + +) to compare and retrieve images having similar compositional elements. ICC + + is an improvement over ICC, specializing in generating low and high-level features (compositional elements) motivated by Max Imdahl's work. To this end, we present a rigorous quantitative and qualitative comparison of our approach with traditional and state-of-the-art (SOTA) methods showing that our proposed method outperforms all of them. In combination with deep features, our method outperforms the best deep learning-based method, opening the research direction for explainable machine learning for digital humanities. We will release the code and the data post-publication. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.