Home Page

Overview

After the success of the past MADiMa workshops, we would like to present to you the MADiMa2024 organized in conjunction with ICPR 2024 the 27th International Conference on Pattern Recognition, Kolkata, India, December 01st, 2024. Kolkata, also known as the ‘City of Joy’, is considered as the cultural capital of India and a historically significant city in the region of Bengal.

The need for accurate, automatic, real-time, and personalised dietary advice has been recently complemented by advances in artificial intelligence (AI) and computer vision (CV), permitting the development of end-to-end pipelines for food content analysis. The development of systems to monitor dietary behavior, analyze food intake, identify eating patterns, and provide feedback and/or suggestions to the user toward healthier nutrition is, hence, a relevant topic to the modern CV community. Estimating food-image content and related nutritional estimation requires overcoming several challenges and stacking several CV modules. In particular, learning-based solutions are generally employed for tasks such as recognition and segmentation of diverse food items. A variety of methods, such as geometry, sensor-based, or neural-based, are employed for 3D modeling and portion estimation. Considering the plethora of challenges present, developing a reliable and fully functional pipeline that consistently delivers accurate results in real-world conditions remains an open area of research. For instance, recognizing food categories in the wild remains an important challenge, given the high inter-class similarities and intra-class variability. Furthermore, the latest advances in Large Language Models (LLMs) and Language Vision Models (LVMs) present a spectrum of new opportunities, e.g., to implement intelligent nutritional assistants and test the actual reliability of these systems in the context of nutritional-content estimation, recipe analysis, food database reading, etc. To this end, the researchers will present and demonstrate their latest progress and discuss novel ideas in the field. Besides the technologies used, emphasis will be given to the precise problem definition, the available nutritional databases, and the evaluation protocols.

Topics

Topics of interest include (but are not limited to) the following:

  • Supervised food recognition.
  • Unsupervised food recognition.
  • Out-of-distribution food detection.
  • Food image synthesis with generative models.
  • Food detection and segmentation.
  • LLMs/LVMs to automatize food composition analysis.
  • Monocular/Binocular depth estimation from mobile/static sensors.
  • 3D point cloud processing and analysis for food volume estimation.
  • Augmented/Virtual Reality for food analysis and portion estimation.
  • Benchmarks, evaluation protocols, and metrics for the above topics.

    • Important Dates

    The submission deadline is at (23:59) 11:59 p.m. of the stated deadline date Anywhere on Earth.


    New paper submission deadline: August 30th September 8th, 2024
    Notification of acceptance:     September 27th, 2024
    Camera-ready deadline:       November 1st, 2024
    Early bird registration deadline:       October 31st, 2024
    Finalized Workshop Program:     October 18th, 2024
    Workshop date: December 01st, 2024

    Contact

    You can reach the workshop chairs though:  info@madima.org