Watch-Bot: Unsupervised learning for reminding humans of forgotten actions

Wu, Chenxia; Zhang, Jiemi; Selman, Bart; Savarese, Silvio; Saxena, Ashutosh

doi:10.1109/icra.2016.7487401

Cited by 15 publications

(9 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This shows the great adaptability property of our framework because the two datasets are quite different in terms of points of view and actions involved. In Table 3, we report the ratio of correctly labeled frames (Accuracy) as in [28,29]. We observe a strong improvement compared to approaches of the literature, i.e.…”

Section: Approachmentioning

confidence: 86%

See 1 more Smart Citation

A New Bayesian Modeling for 3D Human-Object Action Recognition

Maurice

Madrigal

Monin

et al. 2019

2019 16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)

View full text Add to dashboard Cite

HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

show abstract

Section: Approachmentioning

confidence: 86%

“…Accuracy CaTM [28] 38.5 WBTM [29] 41.2 PoT [24] 49.93 KMHIS [11] 59 Figure 7. Qualitative result on the sequence microwaving food on CAD-120.…”

Section: Approachmentioning

confidence: 99%

A New Bayesian Modeling for 3D Human-Object Action Recognition

Maurice

Madrigal

Monin

et al. 2019

2019 16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)

View full text Add to dashboard Cite

show abstract

“…Second, co-occurring actions have variations in temporal orderings, e.g., people can first put-milk-back-to-fridge then microwave-milk instead of the inverse order in the above example, as its ordering is more relevant to the action fetch-milkfrom-fridge. Moreover, these ordering relations could exist in both short-range and long-range, e.g., pouring is followed by drink while sometimes fetch-book is related to put-back-book with a Parts of this work have been published in [57], [58] as the conference version. Fig.…”

Section: Introductionmentioning

confidence: 99%

Watch-n-Patch: Unsupervised Learning of Actions and Relations

Zhang²,

Şener

et al. 2018

IEEE Trans. Pattern Anal. Mach. Intell.

Self Cite

View full text Add to dashboard Cite

There is a large variation in the activities that humans perform in their everyday lives. We consider modeling these composite human activities which comprises multiple basic level actions in a completely unsupervised setting. Our model learns high-level co-occurrence and temporal relations between the actions. We consider the video as a sequence of short-term action clips, which contains human-words and object-words. An activity is about a set of action-topics and object-topics indicating which actions are present and which objects are interacting with. We then propose a new probabilistic model relating the words and the topics. It allows us to model long-range action relations that commonly exist in the composite activities, which is challenging in previous works. We apply our model to the unsupervised action segmentation and clustering, and to a novel application that detects forgotten actions, which we call action patching. For evaluation, we contribute a new challenging RGB-D activity video dataset recorded by the new Kinect v2, which contains several human daily activities as compositions of multiple actions interacting with different objects. Moreover, we develop a robotic system that watches and reminds people using our action patching algorithm. Our robotic setup can be easily deployed on any assistive robots.

show abstract

“…The bag of words paradigm ignores the location of features during feature extraction and attained success in action classification [videos in 7,5]. Most research is based on the classification of the activities after the complete observation of entire video sequence.…”

Section: Figure 1: Taxonomy Of Approaches To Human Action Recognitionmentioning

confidence: 99%

“…In the past five years approach using Spatio Temporal features obtained success in recognizing activities [7]. The bag of words paradigm ignores the location of features during feature extraction and attained success in action classification [videos in 7,5].…”

Section: Figure 1: Taxonomy Of Approaches To Human Action Recognitionmentioning

confidence: 99%

Anticipating Human Activities from Surveillance Videos

Murugesan¹,

Balakrishnan²

2017

ijecs

View full text Add to dashboard Cite

-Human action recognition from the unconstrained surveillance videos and anticipating human action from onset of video is a challenging task. This work focuses on the study of action recognition, action classification followed by action anticipation. We have used the UT Interaction data set containing interactive videos with six types of activities, the system developed classifies three interactive actions punching, pushing and kicking with accuracy. Anticipating action is a probabilistic process and the all possible outcomes are predicted during anticipation. The framework developed can take any interactive video as input from the web camera or any other camera connected to the laptop and will classify the three interactive actions as said above; similarly when the onset of video is given it produces the probable predictions. Action is modeled as a sequence of changes in Spatio Temporal features and histogram of the gradients helps in identifying the changes. The motion boundary descriptors, histogram of oriented gradients and optical flow features are extracted and SVM algorithm is used for classification. Bag of words approach is used for anticipation. Prediction methodologies are highly needed in surveillance environments for efficient monitoring and management.

show abstract

Watch-Bot: Unsupervised learning for reminding humans of forgotten actions

Cited by 15 publications

References 29 publications

A New Bayesian Modeling for 3D Human-Object Action Recognition

A New Bayesian Modeling for 3D Human-Object Action Recognition

Watch-n-Patch: Unsupervised Learning of Actions and Relations

Anticipating Human Activities from Surveillance Videos

Contact Info

Product

Resources

About