The TUM Kitchen Data Set of everyday manipulation activities for motion tracking and action recognition

By Moritz Tenorth, Jan Bandouch, Michael Beetz. 2009

260 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Control and Systems Engineering, Human-Computer Interaction

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 Control and Systems Engineering

In Human-Computer Interaction

Citations in the corpus

Space-Time Representation of People Based on 3D Skeletal Data: A Review (2016, methodology): “6M [92] MoCap color movies 2014 Northwestern-UCLA Multiview [106] Kinect v1 RGB + depth human daily activities 2014 ORGBD [107] Kinect v1 RGB + depth human-object interactions 2014 SPHERE [108] Kinect depth human daily activities 2014 TST Fall Detection [109] Kinect v2 + IMU depth human daily activi”
Towards Understanding Action Recognition (2013, background): “action MPII Cooking [22] y y y TUM kitchen [28] y y y J-HMDB y y y y”
Evaluation of video activity localizations integrating quality and quantity measurements (2014, background): “Consequently, we see an early drop in performance varying any one of the thresholds, t sr = 1 or t sp = 1 .”
A review on vision techniques applied to Human Behaviour Analysis for Ambient-Assisted Living (2012, background): “• TUM kitchen dataset [116]: This dataset targets ADLs at a kitchen scenario at a low action level.”
Hough Forests for Object Detection, Tracking, and Action Recognition (2011, methodology): “We also evaluate our system on four more challenging data sets: the UCF Sports Data Set [80], the UCR Videoweb Activities Data Set [81], the UT-Tower Data Set [82], and the TUM Kitchen Data Set [83].”
Combining embedded accelerometers with computer vision for recognizing food preparation activities (2013, background): “, Darmstadt Daily Routines [10], AmbientKitchen [18], TUM Kitchen [25], CMU-MACC [4], Opportunity [21], ICPR-KSCGR1 and MPII-Cooking [22]).”
A database for fine grained activity detection of cooking activities (2012, background): “Assisted daily living datasets TUM Kitchen [36] det 10 20 4 36,666 384x288 CMU-MMAC [14] cls,det >130 26 1024x768 URADL [20] cls 17 150:30 5 ≤ 50,000 1280x720 Our database cls,det 65 5,609:44 12 881,755 1624x1224”
Does Human Action Recognition Benefit from Pose Estimation? (2011, methodology): “For each type of pose-based feature, we trained a Hough forest on the full 3D skeletons provided (without limb endpoints) [27] as well as a reduced set of 13 joints (see Fig.”
Coupled Action Recognition and Pose Estimation from Multiple Views (2012, methodology): “TUM Kitchen The TUM kitchen dataset (Tenorth et al. 2009) is a more challenging dataset than HumanEva-II.”
A Survey on Human Motion Analysis from Depth Data (2013, background): “Experiments on a multi-view kitchen dataset [41] indicate that the quality of estimated poses with an average error between 42mm-70mm is sufficient for reliable action recognition.”
Ongoing human action recognition with motion capture (2014, methodology): “The training set for these methods is the same as in [17,36,44].”
Efficient Pose-Based Action Recognition (2014, methodology): “The datasets are MSR-Action3D [22], 3D Action Pairs [23], MSRDailyActivity3D(1), and TUM Kitchen dataset [30].”
The Opportunity challenge: A benchmark database for on-body sensor-based activity recognition (2013, background): “The TUM Kitchen dataset focuses on video-based activity recognition (Tenorth et al., 2009), and also contains RFID and reed switch data, but it does not include on-body sensors.”
A Review of Human Activity Recognition Methods (2015, unknown): “…PETS 2004 (CAVIAR) (Fisher, 2004) 2004 6 28 384×288 PETS 2007 (Fisher, 2007b) 2007 3 7 768×576 VIRAT (Oh et al., 2011) 2011 23 17 1,920×1,080 Pose TUM Kitchen (Tenorth et al., 2009) 2009 10 4 20 324×288 Two-person interaction (Yun et al., 2012) 2012 8 7 ≈300 640×480 MSRC-12 Kinect gesture…”
Berkeley MHAD: A comprehensive Multimodal Human Action Database (2013, background): “The TUM Kitchen Dataset by Tenorth et al. [27], which consists of activities in a kitchen setting (i.e., subjects setting a table in different ways), for example includes also RFID tag and magnetic sensor readings in addition to the multi-view video and mocap data.”