Combining embedded accelerometers with computer vision for recognizing food preparation activities

By Sebastian Stein, Stephen J. McKenna. 2013

535 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Electrical and Electronic Engineering, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Electrical and Electronic Engineering

In Signal Processing

Citations in the corpus

Towards Automatic Learning of Procedures From Web Instructional Videos (2017, methodology): “Table 1 shows the comparison between YouCookII and other commonly-used instructional video datasets, e.g., YouCook (Das et al. 2013), MPII (Rohrbach et al. 2012), 50Salads (Stein and McKenna 2013), Breakfast (Kuehne, Arslan, and Serre 2014) and Coffee (Alayrac et al. 2016).”
Contrastive Bidirectional Transformer for Temporal Representation Learning (2019, methodology): “Following [23, 1], we use the five-fold cross-validation resulting in an average of 722 training and 180 test examples.”
RGB-D datasets using microsoft kinect or similar sensors: a survey (2016, background): “3 50 salads dataset School of Computing in University of Dundee created 50 Salads dataset [88] that involves manipulative objects in 2013.”
Scaling Egocentric Vision: The EPIC-KITCHENS Dataset (2018, methodology): “0M 433 3078 50 0 0 52 18 50 Salads [25] × × × 2013 0.”
Temporal Convolutional Networks for Action Segmentation and Detection (2016, methodology): “We use MERL Shopping which was designed for action detection, Georgia Tech Egocentric Activities which was designed for action segmentation, and 50 Salads which has been used for both.”
COIN: A Large-Scale Dataset for Comprehensive Instructional Video Analysis (2019, background): “MPII [35] 9h,48m 44 5,609 cooking activities self-collected - 2012 YouCook [14] 2h,20m 88 cooking activities YouTube - 2013 50Salads [40] 5h,20m 50 966 cooking activities self-collected - 2013 Breakfast [28] 77h 1,989 8,456 cooking activities self-collected 10 2014 "5 tasks" [10] 5h 150 comprehensiv”
Weakly-Supervised Action Segmentation with Iterative Soft Boundary Assignment (2018, background): “Surpassing [15, 18], these two datasets are among the largest datasets for evaluating action segmentation.”
NeuralNetwork-Viterbi: A Framework for Weakly Supervised Video Learning (2018, methodology): “The 50 Salads [29] dataset is another video dataset for action segmentation.”
Temporal Aggregate Representations for Long-Range Video Understanding (2020, methodology): “We experiment on Breakfast [18], 50Salads [34] and EPIC-Kitchens [5].”
RGBD Datasets: Past, Present and Future (2016, background): “Others are more niche: 50 Salads [96] features over 4 hours of people preparing mixed salads.”
Anticipative Video Transformer (2021, methodology): “While well explored in thirdperson video [2, 26, 37, 38, 46, 48, 79, 87], it has recently gained in popularity for first-person (egocentric) videos [13, 14, 16, 24, 55, 62, 74], due to its applicability on wearable computing platforms.”
Predicting the Future from First Person (Egocentric) Vision: A Survey (2021, methodology): “Some relevant third-person datasets which have been used in the literature are: Hollywood2 [100], 50Salads [56], Break-fast [55], ActivityNet-200 [101], OPRA [102].”
LEMMA: A Multi-view Dataset for Learning Multi-agent Multi-task Activities (2020, background): “Conversely, instructional video datasets [1,54,32,31,46] tackle goal-directed multi-step tasks, mostly in cooking, repairing, and assembling activities.”
Vision-based human activity recognition: a survey (2020, unknown): “(b) 50 Salads dataset: It was created at the University of Dundee [188].”
SUN RGB-D: A RGB-D scene understanding benchmark suite (2015, unknown): “There are many existing RGB-D datasets available [54] [47] [1] [25] [44] [60] [49] [45] [29] [66] [57] [52] [46] [16] [21] [73] [35] [67] [3] [41] [10] [63] [42] [64] [65] [48] [12] [33] [8] [50] [26] [6].”
Learning Video Representations using Contrastive Bidirectional Transformer (2019, background): “ on average 5.8 action instances per video. We follow the standard four-fold cross validation split [12, 1] which results in an average of 5,810 training and 1,936 test examples. The 50Salads dataset [23] contains 50 videos with 17 ﬁne-grained action classes required for making a salad, such as “add”