The Language of Actions: Recovering the Syntax and Semantics of Goal-Directed Human Activities

By Hilde Kuehne, Ali Bilgin Arslan, Thomas Serre. 2014

711 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Computational Linguistics & Semantics, Cognitive Science & Psychology

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

1. Luc Van Gool
2. Kristen Grauman
3. Jitendra Malik
4. Cordelia Schmid
5. Andrew Zisserman
6. Trevor Darrell
7. Li Fei-Fei
8. Xiaogang Wang

In Computational Linguistics & Semantics

1. Juergen Gall
2. Alexander Richard
3. Li Fei-Fei
4. Angela Yao
5. Yazan Abu Farha
6. Fadime Sener
7. De-An Huang
8. Juan Carlos Niebles

In Cognitive Science & Psychology

1. Juergen Gall
2. Alexander Richard
3. Li Fei-Fei
4. Angela Yao
5. Yazan Abu Farha
6. Fadime Sener
7. De-An Huang
8. Juan Carlos Niebles

Citations in the corpus

• Towards Automatic Learning of Procedures From Web Instructional Videos (2017, methodology): “Existing datasets for analyzing instructional videos suffer from either limited videos (Alayrac et al. 2016; Sener et al. 2015) or weak diversity in background (Regneri et al. 2013) and activities (Kuehne, Arslan, and Serre 2014)).”
• Rescaling Egocentric Vision: Collection, Pipeline and Challenges for EPIC-KITCHENS-100 (2020, methodology): “Table 4 also compares EPIC-KITCHENS-100 with other datasets used for action anticipation [30,54,55,56,61,62,40].”
• Contrastive Bidirectional Transformer for Temporal Representation Learning (2019, methodology): “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.”
• Hollywood in Homes: Crowdsourcing Data Collection for Activity Understanding (2016, background): “Some relevant big-scale efforts in this direction include MPII Cooking [14], TUM Breakfast [16], and the TACoS Multi-Level [17] datasets.”
• End-to-End Dense Video Captioning with Masked Transformer (2018, neutral): “video lectures, making sandwiches [20], changing tires [1]).”
• Scaling Egocentric Vision: The EPIC-KITCHENS Dataset (2018, methodology): “[24] focuses on cooking dishes based on a list of breakfast recipes.”
• Every Moment Counts: Dense Detailed Labeling of Actions in Complex Videos (2015, methodology): “The MPII Cooking dataset [21] and Breakfast dataset [11] refocus efforts toward human action within restricted action domains (Table 1).”
• Temporal Convolutional Networks for Action Segmentation and Detection (2016, methodology): “[14, 15] modeled actions using Hidden Markov Models on Dense Trajectory features, in tandem with a high-level grammar defining potential action transitions, to 50 Salads.”
• Connectionist Temporal Modeling for Weakly Supervised Action Labeling (2016, methodology): “We follow the metrics used in previous work [19] to evaluate the parsing and segmentation of action units.”
• NeuralNetwork-Viterbi: A Framework for Weakly Supervised Video Learning (2018, methodology): “Compared to the case where all data is available right at the beginning and random shuffling is possible, sorting the 2Each video in the Breakfast dataset belongs to one of ten coarse activities.”
• An end-to-end generative framework for video segmentation and recognition (2015, methodology): “We first compare the accuracy of the proposed reduced FVs against that of our previous work using HTK in combination with HOGHOF for the Breakfast dataset (Table 2).”
• D3TW: Discriminative Differentiable Dynamic Time Warping for Weakly Supervised Action Alignment and Segmentation (2019, methodology): “We follow the metrics used in the previous work [20] to evaluate predicted frame-wise action labels.”
• Neural Task Graphs: Generalizing to Unseen Tasks From a Single Video Demonstration (2018, methodology): “To this end, we evaluate NTG on the few-shot activity classification in the Breakfast Action Dataset [14].”
• Assembly101: A Large-Scale Multi-View Video Dataset for Understanding Procedural Activities (2022, background): “Multi-view fixed-camera datasets include IKEA [2] and Breakfast [22].”
• VideoMamba: State Space Model for Efficient Video Understanding (2024, neutral): “In conclusion, our experiments reveal VideoMamba’s potential in understanding both short-term (K400 [37] and SthSthV2 [24]) and long-term (Breakfast [38], COIN [72], and LVU [86]) video contents.”
• Temporal Aggregate Representations for Long-Range Video Understanding (2020, background): “Notable works include two steam networks [31,40], 3D convolutional networks [36,3], and RNNs [7,44].”