Counting Bites and Recognizing Consumed Food from Videos for Passive Dietary Monitoring

نویسندگان

چکیده

Assessing dietary intake in epidemiological studies are predominantly based on self-reports, which subjective, inefficient, and also prone to error. Technological approaches therefore emerging provide objective assessments. Using only egocentric videos, this work aims accurate estimation individual through recognizing consumed food items counting the number of bites taken. This is different from previous that rely inertial sensing count bites, recognize visible but not ones. As a subject may consume all meal, those more valuable. A new dataset has 1,022 video clips was constructed validate our concept bite item recognition videos. 12 subjects participated 52 meals were captured. total 66 unique items, including ingredients drinks, labelled along with 2,039 bites. Deep neural networks used perform an end-to-end manner. Experiments have shown directly can reach 74.15% top-1 accuracy (classifying between 0-4 20-second clips), MSE value 0.312 (when using regression). Our experiments video-based show indeed harder than ones, drop 25% F1 score.

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Recognizing Human Faces from Videos

We present an exemplar-based probabilistic approach for face recognition in video data. The approach has two stages: First, Exemplars, which are selected representatives from the raw video, are automatically extracted from gallery videos. The exemplars are used to summarize the gallery video information. In the second part, these exemplars are then used as centers for probabilistic mixture dist...

متن کامل

Unsupervised Feature Learning from Videos for Discovering and Recognizing Actions

In this work, we evaluate different unsupervised feature learning approaches using Convolutional Neural Networks (CNNs) and videos. Essentially, we experiment with a CNN feature learning model based on the Contrastive Loss function. We propose a novel Siamese-tuple network architecture trained with a new loss function which benefits from the temporal coherence present in videos as the unique fo...

متن کامل

Recognizing Human Activities from Partially Observed Videos

Recognizing human activities in partially observed videos is a challenging problem and has many practical applications. When the unobserved subsequence is at the end of the video, the problem is reduced to activity prediction from unfinished activity streaming, which has been studied by many researchers. However, in the general case, an unobserved subsequence may occur at any time by yielding a...

متن کامل

Tracking Objects from Multiple Soccer Videos and Recognizing Events

This paper presents a novel way of recognizing events from a soccer match video sequence. After tracking players and the ball, events such as passing, kicking, having, scoring and struggling for the ball can be inferred for the subsequences. In our framework, the ball trajectory is represented as an image blob broken with cuts caused by occlusion or interference from a player. For those frame c...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: IEEE Journal of Biomedical and Health Informatics

سال: 2021

ISSN: ['2168-2208', '2168-2194']

DOI: https://doi.org/10.1109/jbhi.2020.3022815