Food volume estimation for quantifying dietary intake with a wearable camera

Gao, Anqi; Lo, Frank P.-W.; Lo, Benny

doi:10.1109/bsn.2018.8329671

Cited by 34 publications

(27 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1(c). If a video is captured, a real-time 3-D reconstruction technique, which has been proposed in [6], is used as an alternative choice to reconstruct the partial point cloud with more 3-D information compared to that using a single depth image. Fourth, the partial point cloud is then directed to the point completion network to perform 3-D reconstruction and estimate the portion size of the food items, as shown in Fig.…”

Section: Detailed Information and Methodsmentioning

confidence: 99%

“…The difference between SfM-based and SLAM-based 3-D reconstruction techniques is that the SLAMbased approach estimates camera motion and reconstructs 3-D models in real-time. In [6], the authors developed a real-time 3-D reconstruction method to estimate the food portion size. This proposed technique can achieve around 83% accuracy examined with similar food types captured in the wild.…”

Section: A Volume Estimation Approachesmentioning

confidence: 99%

“…After a comprehensive exploration, we found that an integrated approach based on deep learning and 3-D reconstruction could be one of the potential solutions in aiding dietary assessment. Specifically, deep learning view synthesis [14] can be used along with the SLAM-based approach [6] to estimate the volume of food items without the need for the users to shift and rotate the camera to obtain the complete 360 • view of the food. Although a number of research works, such as PointNet [15] and PointNet++ [16], have explored the efficacy of using raw point cloud for classification, there are relatively fewer works using point cloud to perform view synthesis.…”

Section: Deep Learning View Synthesis On 3-d Modelsmentioning

confidence: 99%

“…To estimate the portion size, however, it relies heavily on individuals' subjective perception, which could be highly biased and inaccurate. It is for these reasons that various objective vision-based methods, ranging from model-based [4], [5], stereo-based [6], depth camera based [7], and deep learning approaches [8], have been proposed. While these approaches present reasonable accuracy in portion size estimation, there still exist several key challenges such as view occlusion and scale ambiguity.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Point2Volume: A Vision-Based Dietary Assessment Approach Using View Synthesis

Sun

Qiu

et al. 2020

IEEE Trans. Ind. Inf.

Self Cite

View full text Add to dashboard Cite

Dietary assessment is an important tool for nutritional epidemiology studies. To assess the dietary intake, the common approach is to carry out 24-h dietary recall (24HR), a structured interview conducted by experienced dietitians. Due to the unconscious biases in such self-reporting methods, many research works have proposed the use of vision-based approaches to provide accurate and objective assessments. In this article, a novel vision-based method based on real-time three-dimensional (3-D) reconstruction and deep learning view synthesis is proposed to enable accurate portion size estimation of food items consumed. A point completion neural network is developed to complete partial point cloud of food items based on a single depth image or video captured from any convenient viewing position. Once 3-D models of food items are reconstructed, the food volume can be estimated through meshing. Compared to previous methods, our method has addressed several major challenges in vision-based dietary assessment, such as view occlusion and scale ambiguity, and it outperforms previous approaches in accurate portion size estimation.

show abstract

Section: Detailed Information and Methodsmentioning

confidence: 99%

Section: A Volume Estimation Approachesmentioning

confidence: 99%

Section: Deep Learning View Synthesis On 3-d Modelsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Point2Volume: A Vision-Based Dietary Assessment Approach Using View Synthesis

Sun

Qiu

et al. 2020

IEEE Trans. Ind. Inf.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Food image analysis can provide an accurate tool for monitoring of ingestive behavior by capturing imagery of food intake. Food identification using deep learning methods [15], [16] and portion size estimation models [17], [18] were reported as methods of food image assessment. The progressive trend towards a food image-based dietary assessment has received much focus on several imaging methods that can be employed in dietary assessment.…”

Section: Introductionmentioning

confidence: 99%

Detection of Oil-Containing Dressing on Salad Leaves Using Multispectral Imaging

Raju

Sazonov

2020

IEEE Access

View full text Add to dashboard Cite

There is a need for non-invasive methods for nutritional analysis of food that can address the drawbacks of current methods such as color photography, which cannot distinguish between energydense and zero calorie foods. This paper discusses a novel, multispectral approach for the identification of oil, particularly in salads, and defines a pseudo-reflectance term. A custom-made multispectral camera was used to collect a novel, publicly shared dataset of images of untreated lettuce leaves or leaves treated with vinegar, oil, or a combination of these. The camera captured image data at 10 wavelengths ∈[380nm,980nm] across the electromagnetic spectrum in the visible and NIR (near-infrared) regions. Imaging was done in a lab environment with the presence of ambient light. Mean spectra were extracted from the regions of interest in the multispectral cube and used to compute pseudo-reflectance. ANOVA (Analysis of Variance) was performed to look for variances in the pseudo-reflectance curves. ANOVA proved that the differences between group means of the four treatment groups (oil, vinegar, oil plus vinegar, control) were statistically significant. Pairs of groups showing the greatest significance were established using a Tukey post hoc test. Sequential Forward Selection (SFS) was used to determine 5 optimal feature wavelengths from the whole feature space (410 nm, 455 nm, 485 nm, 810 nm, and 850 nm). A combination of visible (VIS) and infrared (IR) wavelengths, selected using SFS, showed the greatest potential for discrimination between groups containing oil and groups that do not contain oil with a classification accuracy of 84.20%. The pseudo-reflectance values were statistically proven to be sensitive to the presence of oil as a dressing. This research has demonstrated the feasibility of implementing a multispectral imaging technique for identifying the presence of oil in salads and possibly an energy content detection system. INDEX TERMS Multispectral imaging, energy density, food detection, high calorie detection, dietary assessment.

show abstract