Marc Bolaños

Home Page

I am Marc Bolaños, a PhD Candidate at the CVUB group of the Universitat de Barcelona under the supervision of professor Petia Radeva.

I am working on the development of Computer Vision and Deep Learning methodologies applied on Egocentric images for the analysis and storytelling of the daily life of the users and also on Food analysis for the improvement of healthy eating habits.

Latest Projects

logmeal api food analysis

LOGMEAL

LogMeal Demo API for Food Analysis.

Online Demo
Bolaños, M., Aguilar, E., Soriano J., Remeseiro, B., & Radeva, P.
food ingredients recognition

FOOD INGREDIENTS RECOGNITION

Food Ingredients Recognition through Multi-label Learning.

Dataset Recipes5k Dataset Ingredients101
Bolaños, M., Ferrà, A., & Radeva, P. (2017). 
“Food Ingredients Recognition through Multi-label Learning”. 
In 3rd International Workshop on Multimedia Assisted Dietary Management, ICIAP. (In press)
egocentric description temporal

EGOCENTRIC TEXTUAL DESCRIPTION

Temporally-linked Multi-input Attention (TMA) model proposed for generating natural language descriptions for egocentric sequences.

Code Dataset
Bolaños, M., Peris, Á., Casacuberta, F., Soler, S., and Radeva, P. (2017).
"Egocentric Video Description based on Temporally-Linked Sequences".
In Special Issue on Egocentric Vision and Lifelogging Tools.
Journal of Visual Communication and Image Representation (VCIR), (SUBMITTED).
food localization and recognition

FOOD LOCALIZATION AND RECOGNITION

Simultaneous Food Localization and Recognition on both conventional and egocentric images.

Dataset
Bolaños, M., and Radeva, P. (2016)
“Simultaneous Food Localization and Recognition”
In 23rd International Conference on Pattern Recognition (ICPR)
video description

ABiViRNet FOR VIDEO DESCRIPTION

ABiViRNet: Attention Bidirectional Video Recurrent Net, model for video captioning.

Code
Peris, Á., Bolaños, M., Radeva, P., and Casacuberta, F. (2016)
“Video Description using Bidirectional Recurrent Neural Networks”
In Proceedings of the 25th International Conference on Artificial Neural Networks (ICANN)
vibiknet

VIBIKNet FOR VISUAL QUESTION ANSWERING

Visual Bidirectional Kernelized Network for Visual Question Answering.

Code
Bolaños, M., Peris, Á., Casacuberta, F., & Radeva, P.
“VIBIKNet: Visual Bidirectional Kernelized Network for the VQA Challenge”
VQA Challenge, CVPR '16 (No Proceedings)
sr-clustering

SEMANTIC R-CLUSTERING

Unsupervised algorithm for Segmentation in Events of Egocentric Vision photo streams. It uses a rich frames representation of both semantic and global features extracted by means of Convolutional Neural Networks.

Code Dataset
Dimiccoli, M., Bolaños, M., Talavera, E., Aghaei, M., Nikolov, S. & Radeva, P. (2015)
"SR-Clustering: Semantic Regularized Clustering for Egocentric Photo Streams Segmentation".
Pre-print: http://arxiv.org/abs/1512.07143

EGO-OBJECT DISCOVERY

Semi-supervised and iterative algorithm for Object Discovery on Egocentric Images.

Code Dataset
Bolaños, M. & Radeva, P. (2015).
“Ego-object discovery”.
Pre-print: http://arxiv.org/abs/1504.01639

SEMANTIC SUMMARIZATION

Summarization of Egocentric Events based on an initial CNN-based filtering, a semantic relevance ranking and a final diversity re-ranking for offering a diverse set of keyframes.

Caffe Model
Lidon, A., Bolaños, M., Dimiccoli, M., Radeva, P., Garolera, M. and Giró-i-Nieto, X. (2015).
“Semantic Summarization of Egocentric Photo Stream Events”.
Pre-print: http://arxiv.org/abs/1511.00438

MOTION-BASED SEGMENTATION

Egocentric Vision Event Segmentation based on robust motion SIFT-Flow features for low temporal resolution photo streams (2-3 fpm).

Code
Bolaños, M., Garolera, M., & Radeva, P. (2014).
"Video segmentation of life-logging videos".
In Articulated Motion and Deformable Objects (pp. 1-9). Springer International Publishing.

Publications

Journal Papers

Bolaños, M., Peris, Á., Casacuberta, F., Soler, S. & Radeva, P. (2017). "Egocentric Video Description based on Temporally-Linked Sequences". In Special Issue on Egocentric Vision and Lifelogging Tools. Journal of Visual Communication and Image Representation (VCIR), (SUBMITTED). PDF
Dimiccoli, M., Bolaños, M., Talavera, E., Aghaei, M., Nikolov, S. & Radeva, P. (2015) "SR-Clustering: Semantic Regularized Clustering for Egocentric Photo Streams Segmentation". In Computer Vision and Image Understanding (CVIU) (pp. 55-69). Elsevier. PDF
Bolaños, M., Dimiccoli, M. & Radeva, P. (2015). “Toward storytelling from visual lifelogging: An overview”. In Special Issue on Wearable and Ego-Vision Systems for Augmented Experience, IEEE Transactions on Human-Machine Systems (THMS) 47 (1), 77–90. PDF
Bolaños, M. & Radeva, P. (2015). “Ego-Object Discovery”. PDF
Presentation

Conference Papers

Aguilar, E., Bolaños, M., and Radeva, P. (2017). “Food Recognition using Fusion of Classifiers based on CNNs”. In International Conference of Image Analysis and Processing (ICIAP) (in press).
Aguilar, E., Bolaños, M., and Radeva, P. (2017). “Exploring Food Detecting using CNNs”. Eurocast’2017.
Bolaños, M., Peris, Á., Casacuberta, F., & Radeva, P. “VIBIKNet: Visual Bidirectional Kernelized Network for Visual Question Answering” Iberian Conference on Pattern Recognition and Image Analysis, IbPRIA '17. PDF
Bolaños, M. and Radeva, P. (2016). “Simultaneous Food Localization and Recognition on Egocentric Images”. In Proceedings of the 23rd International Conference on Pattern Recognition (ICPR). PDF
Poster
Herruzo, P., Bolaños, M. and Radeva, P. (2016). “Can a CNN Recognize Catalan Diet?”. In Proceedings of the 8th Intl Conf. for Promoting the Application of Mathematics in Technical and Natural Sciences (AMiTaNS). PDF
Peris, Á., Bolaños, M., Radeva, P., and Casacuberta, F. (2016). “Video Description using Bidirectional Recurrent Neural Networks”. In Proceedings of the 25th International Conference on Artificial Neural Networks (ICANN). PDF
Marone, J., Balocco, S., Bolaños, M., Massa, JM., and Radeva, P. (2016). “Learning the Lumen Border using a Convolutional Neural Networks classifier”. In Proceedings of the 19th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI).
Talavera, E., Dimiccoli, M., Bolaños, M., Aghaei, M., & Radeva, P. (2015). “R-Clustering for egocentric video segmentation”. In Pattern Recognition and Image Analysis (pp. 327-336). Springer International Publishing. Poster
Bolaños, M., Garolera, M., & Radeva, P. (2015). “Object Discovery Using CNN Features in Egocentric Videos”. In Pattern Recognition and Image Analysis (pp. 67-74). Springer International Publishing. Presentation
Bolaños, M., Garolera, M., & Radeva, P. (2014). “Video Segmentation of Life-Logging Videos”. In Articulated Motion and Deformable Objects (AMDO) (pp. 1-9). Springer International Publishing.

Challenges

Bolaños, M., Peris, Á., Casacuberta, F., & Radeva, P. “VIBIKNet: Visual Bidirectional Kernelized Network for the VQA Challenge” VQA Challenge, CVPR '16 (No Proceedings) Poster
Lidon, A., Bolaños, M., Seidl, M., Giró-i-Nieto, X., Radeva, P., & Zeppelzauer, M. (2015, August). “UPC-UB-STP@ MediaEval 2015 Diversity Task: Iterative Reranking of Relevant Images”. In Retrieving Diverse Social Images Task, MediaEval 2015.
de Oliveira Barra, G., Ayala, A. C., Bolaños, M., Dimiccoli, M., Giro-i-Nieto, X., & Radeva, P. “LEMoRe: A Lifelog Engine for Moments Retrieval at the NTCIR-Lifelog LSAT Task”. Age, 40(33), 48.

Workshop Papers

Bolaños, M., Ferrà, A., & Radeva, P. (2017) “Food Ingredients Recognition through Multi-label Learning” In 3rd International Workshop on Multimedia Assisted Dietary Management, ICIAP. (in press) PDF
Oliveira-Barra, G., Bolaños, M., Talavera, E., Dueñas, A., Gelonch, O. & Garolera, M. (2017) “Serious Games Application for Memory Training Using Egocentric Images” In 3rd International Workshop on Multimedia Assisted Dietary Management, ICIAP. (in press) PDF
Lidon, A., Bolaños, M., Dimiccoli, M., Radeva, P., Garolera, M. & Giró-i-Nieto, X. (2015) "Semantic Summarization of Egocentric Photo Stream Events". (SUBMITTED) PDF
Bolaños, M., Mestre, R., Talavera, E., Giró-i-Nieto, X. & Radeva, P. (2015). “Visual Summary of Egocentric Photostreams by Representative Keyframes”. In International Workshop on Wearable and Ego-vision Systems for Augmented Experience (WEsAX). ICMEW. PDF
Presentation
Bolaños, M., Garolera, M., & Radeva, P. (2013, October). “Active labeling application applied to food-related object recognition”. In Proceedings of the 5th international workshop on Multimedia for cooking & eating activities (pp. 45-50). ACM. PDF

Contact

  • Address

    Gran Via de les Corts Catalanes, 585. 08007, Barcelona. Spain
  • Phone

    +34 93 402 18 97
  • Email

    marc(dot)bolanos(at)ub(dot)edu