Semantic Art Understanding
In this project, we aim to find textual and visual embeddings for understanding images of fine-art paintings in a semantic way.
- Noa Garcia, George Vogiatzis. How to Read Paintings: Semantic Art Understanding with Multi-Modal Retrieval. European Conference in Computer Vision Workshops, 2018.
This project was part of my PhD thesis. Here, we use query images to find and retrieve segments of videos in large-scale video collections.
- Noa Garcia, George Vogiatzis. Asymmetric Spatio-Temporal Embeddings for Large-Scale Image-to-Video Retrieval. British Machine Vision Conference, 2018.
- Noa Garcia. Temporal Aggregation of Visual Features for Large-Scale Image-to-Video Retrieval. ACM Internation Conference on Multimedia Retrieval, 2018.
- Noa Garcia, George Vogiatzis. Dress like a Star: Retrieving Fashion Products from Videos. IEEE International Conference on Computer Vision Workshop, 2017.
This project addresses the problem of detecting and classifying parking spaces in either car parks or in the street using 3D reconstruction techniques and feature analysis. The project was conducted at G@TV group at Universidad Politecnica de Madrid and it was funded by the Spanish Industry, Tourism and Commerce Council and the ERDF (reference TSI-020100-2011-57).
- Nuria Sanchez, Noa Garcia, Jose Manuel Menendez. Feedback-Based Parameterized Strategies for Improving Performance of Video Surveillance Understanding Frameworks. Ibero-American Conference on Artificial Intelligence, 2014.
- Carmen Bravo, Nuria Sanchez, Noa Garcia, Jose Manuel Menendez. Outdoor Vacant Parking Space Detector for Improving Mobility in Smart Cities. Portuguese Conference on Artificial Intelligence, 2013.
Heart Rate Estimation
My first computer vision related project conducted at Gradiant as my final undergrad project. We implemented a system to estimate the heart rate of a person by using facial video information.
- Noa Garcia Docampo. Heart rate estimation using facial video information. Universitat Politecnica de Catalunya, 2012.