Pierre-Etienne Martin
Postdoctoral Researcher & Tech Development Coordinator
Abteilung für Vergleichende Kulturpsychologie
Max-Planck-Institut für evolutionäre Anthropologie
Deutscher Platz 6
04103 Leipzig
Telefon: +49 (0) 341 3550 460
E-Mail:
pierre_etienne_martin@[>>> Please remove the text! <<<]eva.mpg.de
Research Interests
- Image Processing and Computer Vision
- Fine grained action classifications
- Deep learning methods
- Optical Flow analysis
- Feature analysis
Current Projects
Currently Postdoctoral Researcher & Tech Development Coordinator at the Max Planck Institute for Evolutionary Anthropology in the Comparative Cultural Psychology department, I apply and develop computer vision tools with the aim of better understanding the human and non-human animal mind. We also aim to change our way of acquiring data by encouraging non-invasive methods. We hope to strengthen hypotheses and conclusions through big data analysis and model robustness.
Zoo Cam Set-Up
We aim to build a portable setup to locate, track and identify non-human primates in their enclosure or natural environment. Such a project can be divided into several independent tasks:
individual detection & tracking
track-line classification
3D localization from several images
In order to retrieve qualitative information, the Zoo Cam Set-Up must have synchronized high-resolution images and static cameras.
BioTIP - Bio-Signal Retrieval from Thermal Imaging Processing
Our focus is on building a set of computer vision tools for the psychology and animal behavior research communities to retrieve bio-signals in an ecological manner. We aim to increase the use of thermal imaging modality in the community and avoid using more invasive recording methods to answer research questions.
We created the ApeNose dataset, constituted of TI images of chimpanzees with their face (bbox) and nose (9 landmarks) annotated. Several state-of-the-art methods are then compared with our implemented Tifa (Face) and Tina (Nose) models. We aim to analyze the temporal evolution of certain regions and landmarks in order to retrieve physiological signals that help us better understand the human and non-human mind.
Quantex - Quantifying Lived Experience
Quantex focuses on the analysis of lived experience in childhood. For this purpose, families participate in our acquisition project by letting their child wear a vest equipped with a camera. In order to process and understand these egocentric videos, different computer vision challenges remain to be solved:
object interaction detection
person detection and identification
action localization and classification
After have investigated the automatic detection of interaction with objects, we are focusing on the type of activities the child is performing alone or with others.
CASE - Computer-Animal Self-testing Environment
CASE aims to allow non-human animal individuals to test themselves independently in their enclosure at their own speed over several days. The state of the experiment should be saved when the individual leaves and loaded back when needed. Data acquisition can therefore be conducted without human intervention. The data may then be collected and processed remotely, as the experiment can be changed or updated. A dataset has been automatically generated using recordings from ZACI devices, automatic detection of individuals and low effort annotations. This dataset can be derived in several versions, leading to different performances for individual classification. This dataset is currently being enriched for further investigation.
Curriculum Vitae
Current Positions
2021 - present | Postdoctoral Researcher & Tech Development Coordinator |
2021 - 2022 | Scientific Advisor at DeepMove, Bordeaux, France |
Career
2020 - 2021 | ATER Temporary Teaching and Research Associate at LaBRI University of Bordeaux, Talence, France |
2017 - 2020 | PhD student at LaBRI University of Bordeaux, Talence, France |
2017 | Master Internship Institute of Thermophysics, Novosibirsk, Russia |
Education
2021 | French academic qualifications in Section 27 and 61 Computer Science & engineering, automation and signal processing |
2017 - 2020 | PhD in Computer Science, LaBRI University of Bordeaux, France Thesis: Fine-grained action detection and classification from videos with spatio-temporal convolutional neural networks. Application to Table Tennis. Supervisors: Jenny Benois Pineau - LaBRI, University of Bordeaux; Renaud Péteri - MIA, University of La Rochelle |
2015 - 2017 | M.Sc. Erasmus Mundus Master in Image Processing & Computer Vision PPCU Budapest, Hungary Universidad Autonoma de Madrid, Spain University of Bordeaux, France |
2014 - 2015 | B.Sc. in Mathematical Engineering University of Bordeaux, France |
2011 - 2013 | Preparatory Class for Prestigious Engineering Schools in Physics and Chemistry Lycée Camille Julian, Bordeaux, France |
Teaching
2017 - 2021 | as PhD student & ATER in LaBRI |
Master classes:
- Deep Learning in Computer Vision
- Artificial Intelligence
- Data Analysis, Classification, Indexing
Bachelor classes:
- Fundamental Algorithms and Basis of Programming
- Excel, VBA
- Excel, Algorithms, Python
- Databases and Web Programming
- Table Algorithms
- Digital Culture and Skills
- Computer and internet certificate
Public outreach
- SAI Interview in 2023 - Video
- Climb the wall in 2023 - Article
- MediaEval Sport Task presentations from 2019 to 2023 - Youtube channel
- Ma thèse en 1024 caractères in 2020 - Article
- Ma thèse en 180 secondes in 2019 - MT180
- Jamming Assembly in 2019 - Video
- CNRS Village in 2019 - Picture
- The European Researchers’ Night in 2018 and 2019 - Website
- Discussion between Researchers and Secondary School Students (Declics) in 2018 - Website
Pierre-Etienne Martin, Jenny Benois-Pineau, Renaud Péteri, Julien Morlier. Fine grained sport action recognition with Twin spatio-temporal convolutional neural networks. Multimedia Tools and Applications, Springer Verlag, 2020.
DOI hal-02551019
Buchkapitel
Martin, P.-E. (2024). Dataset generation and Bonobo classification from weakly labelled videos. In K. Arai ( |
|
Zahra, A., Martin, P.-E., Bohn, M., & Haun, D. (2024). Computer vision for analyzing children’s lived experiences. In K. Arai ( |
|
Martin, P.-E., Benois-Pineau, J., Péteri, R., Zemmari, A., & Morlier, J. (2021). 3D convolutional networks for action recognition: Application to sport gesture recognition. In J. Benois-Pineau, & A. Zemmari ( |
Konferenzbeiträge
Hacker, L., Bartels, F., & Martin, P.-E. (2023). Fine-grained action detection with RGB and pose information using two stream convolutional networks. In CEUR Workshop Proceedings: MediaEval 2022: Multimedia Evaluation Workshop. |
|
Martin, P.-E. (2023). Baseline method for the Sport Task of MediaEval 2022 with 3D CNNs using attention mechanisms. In CEUR Workshop Proceedings: MediaEval 2022: Multimedia Evaluation Workshop. |
|
Martin, P.-E., Calandre, J., Mansencal, B., Benois-Pineau, J., Péteri, R., Mascarilla, L., & Morlier, J. (2023). Sport Task: Fine grained action detection and classification of table tennis strokes from videos for MediaEval 2022. In CEUR Workshop Proceedings: MediaEval 2022: Multimedia Evaluation Workshop. |
|
Martin, P.-E. (2021). Spatio-temporal CNN baseline method for the sports video task of MediaEval 2021 benchmark. In Proceedings MediaEval 2021: Multimedia Benchmark Workshop. |
|
Martin, P.-E., Calandre, J., Mansencal, B., Benois-Pineau, J., Mascarilla, R. P. L., & Morlier, J. (2021). Sports video: Fine-grained action detection and classification of table tennis strokes from videos for MediaEval 2021. In Proceedings MediaEval 2021: Multimedia Benchmark Workshop. |
|
Zahra, A., & Martin, P.-E. (2021). Two stream network for stroke detection in table tennis. In Proceedings of MediaEval 2021: Multimedia Benchmark Workshop. |
|
Martin, P.-E., Benois-Pineau, J., Péteri, R., & Morlier, J. (2021). Three-stream 3D/1D CNN for fine-grained action classification and segmentation in table tennis. In MMSports'21: Proceedings of the 4th International Workshop on Multimedia Content Analysis in Sports (pp. 35-41). |
Pierre-Etienne Martin, Jenny Benois-Pineau, Renaud Péteri, Julien Morlier. 3D attention mechanism for fine-grained classification of table tennis strokes using a Twin Spatio-Temporal Convolutional Neural Networks. 25th International Conference on Pattern Recognition (ICPR2020), Jan 2021, Milano, Italy.
hal-02977646
Pierre-Etienne Martin, Jenny Benois-Pineau, Boris Mansencal, Renaud Péteri, Laurent Mascarilla, et al.. Sports Video Classification: Classification of Strokes in Table Tennis for MediaEval 2020. MediaEval 2020 Workshop, Jan 2021, Online, Unknown Region.
hal-03104270
Pierre-Etienne Martin, Jenny Benois-Pineau, Boris Mansencal, Renaud Péteri, Julien Morlier. Classification of Strokes in Table Tennis with a Three Stream Spatio-Temporal CNN for MediaEval 2020. MediaEval 2020 Workshop, Dec 2020, Online, Unknown Region.
hal-03104275
Kazi Ahmed Asif Fuad, Pierre-Etienne Martin, Romain Giot, Romain Bourqui, Jenny Benois-Pineau, et al.. Features Understanding in 3D CNNs for Actions Recognition in Video. Tenth International Conference on Image Processing Theory, Tools and Applications, IPTA 2020, Oct 2020, Paris, France.
hal-02963298
Pierre-Etienne Martin, Jenny Benois-Pineau, Renaud Peteri, Julien Morlier. Optimal Choice of Motion Estimation Methods for Fine-Grained Action Classification with 3D Convolutional Networks. 2019 IEEE International Conference on Image Processing (ICIP), Sep 2019, Taipei, Taiwan. pp.554-558.
10.1109/ICIP.2019.8803780 hal-02326240
Pierre-Etienne Martin, Jenny Benois-Pineau, Renaud Péteri. Fine-Grained Action Detection and Classification in Table Tennis with Siamese Spatio-Temporal Convolutional Neural Network. 2019 IEEE International Conference on Image Processing (ICIP), Sep 2019, Taipei, Taiwan. pp.3027-3028.
10.1109/ICIP.2019.8803382 hal-02326229
Pierre-Etienne Martin, Jenny Benois-Pineau, Boris Mansencal, Renaud Péteri, Jordan Calandre, et al.. Sports Video Annotation: Detection of Strokes in Table Tennis task for MediaEval 2019. MediaEval 2019 Workshop, Oct 2019, Sophia Antipolis, France.
hal-02937666
Pierre-Etienne Martin, Jenny Benois-Pineau, Boris Mansencal, Renaud Péteri, Julien Morlier. Siamese Spatio-temporal convolutional neural network for stroke classification in Table Tennis games. MediaEval 2019 Workshop, Oct 2019, Sophia Antipolis, France.
hal-02937668
Pierre-Etienne Martin, Jenny Benois-Pineau, Renaud Péteri, Julien Morlier. Sport Action Recognition with Siamese Spatio-Temporal CNNs: Application to Table Tennis. 2018 International Conference on Content-Based Multimedia Indexing (CBMI), Sep 2018, La Rochelle, France. pp.1-6,
10.1109/CBMI.2018.8516488 hal-02360011
PhD Thesis
Pierre-Etienne Martin. Fine-Grained Action Detection and Classification from Videos with Spatio-Temporal Convolutional Neural Networks. Application to Table Tennis.. Neural and Evolutionary Computing [cs.NE]. Université de Bordeaux; Université de la Rochelle, 2020. English.
tel-03099907
Master Thesis
Pierre-Etienne Martin. Face recognition using depth information. Master thesis, Institute of Thermophysics of Novosibirsk, Russia, 2017.