Guide to the Carnegie Mellon University Multimodal Activity (CMU-MMAC) Database

By Fernando De la Torre, Jessica K. Hodgins, Adam W. Bargteil, Xavier Martin, J. Macey, A. Collado, Pep Beltran. 2008

322 citations recorded by the source corpus.

Subfields: Multimodal Learning, Evaluation & Benchmarks

How researchers see this paper

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

In Multimodal Learning

In Evaluation & Benchmarks

Citations in the corpus

Ego4D: Around the World in 3,000 Hours of Egocentric Video (2021, background): “Whereas today’s largest ego datasets focus solely on kitchens [44,44,124,138], Ego4D spans hundreds of environments both indoors and out-doors.”
Scaling Egocentric Vision: The EPIC-KITCHENS Dataset (2018, methodology): “2 Related Datasets We compare EPIC-KITCHENS to four commonly-used [6, 13, 36, 8] and two recent [16, 41] egocentric datasets in Table 1, as well as six third-person activityrecognition datasets [14, 42, 56, 28, 44, 39] that focus on object-interaction activities.”
On Human Action (2011, background): “This, however, is likely to be necessary in scenarios that are highly variable such as our kitchen scenario which is indeed being explored by work such as [2] and examples of which are contained the “kitchen dataset” [9].”
A review on vision techniques applied to Human Behaviour Analysis for Ambient-Assisted Living (2012, background): “• CMU–MMAC database [26]: The multi-modal activity database from the Carnegie Mellon University targets cooking and food preparation activities.”
Fuzzy human motion analysis: A review (2014, unknown): “CMU MoCap 2008 [208] [152] RA1⁄498.”
Berkeley MHAD: A comprehensive Multimodal Human Action Database (2013, background): “[7] contains multimodal measures captured from subjects performing tasks such as meal preparation and cooking.”
Towards Understanding Action Recognition (2013, background): “While marker-based motion capture systems provide accurate 3D ground-truth pose data [12, 15, 19, 25], they are impractical for recording realistic video data.”
Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectives (2023, unknown): “…understanding, facilitated by recent ego-video datasets showing unscripted daily-life activity as in Ego4D [47], EPIC-Kitchens [27, 28, 163], UT Ego [78], ADL [119], and KrishnaCam [147], or procedural activities as in EGTea [81], AssistQ [172], Meccano [126], CMU-MMAC [77], and EgoProcel [10].”
TailorNet: Predicting Clothing in 3D as a Function of Human Pose, Shape and Garment Style (2020, background): “These models [10, 18, 17] tend to produce over-smooth results, with the exception of [29], but the model is trained for a single garment and subject, and as mentioned before generalization to in the wild motions (for example CMU [11]) is not demonstrated.”
The EPIC-KITCHENS Dataset: Collection, Challenges and Baselines (2020, background): “[21], [22], [23], [24], [25] have been significantly smaller in size”
Summarization of Egocentric Videos: A Comprehensive Survey (2017, methodology): “Even if recorded with head-mounted devices, this is the case of datasets such as the following: CMU-MMAC [51], recorded in a staged kitchen; GTEA [52] and GTEA-gaze [50], which contain very specific videos, sometimes staged and short; UEC [53], which is a compilation of short videos from YouTube and”
Predicting the Future from First Person (Egocentric) Vision: A Survey (2021, methodology): “The CMU dataset [44] captured human-object interactions by using RF-ID tags.”
Interpreting and Executing Recipes with a Cooking Robot (2012, unknown): “First, we compared our supported primitives to a brownie recipe performed by a human, annotated within the CMU Multimodal Activity Database [7].”
USC-HAD: a daily activity dataset for ubiquitous activity recognition using wearable sensors (2012, background): “The Carnegie Mellon University Multi-Modal Activity Database (CMU-MMAC) is different from the datasets mentioned above in the sense that it contains many other modalities besides accelerometers and gyroscopes to sense and measure human activities [21].”