Pose Tutor: An Explainable System for Pose Correction in the Wild

Dittakavi, Bhat; Bavikadi, Divyagna; Desai, Sai Vikas; Chakraborty, Soumi; Reddy, Nishant; Balasubramanian, Vineeth N; Callepalli, Bharathi; Sharma, Ayon

doi:10.1109/cvprw56347.2022.00398

Cited by 35 publications

(12 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…An angle detection mechanism was used for pose predictions and to detect wrongly formed joints. [ 19 ] TAGteach, an acoustic guiding system involving auditory feedback such as the generation of sound when the desired behavior occurs has been implemented to correct posture in sports, dance, and, walking. However, no correction procedure has been discussed in their work.…”

Section: Literature Reviewmentioning

confidence: 99%

Smart Yoga Instructor for Guiding and Correcting Yoga Postures in Real Time

Kishore

Bindu²,

Manjunath

2022

International Journal of Yoga

View full text Add to dashboard Cite

In recent days, Yoga is gaining more prominence and people all over the world have started to practice it. Performing Yoga with proper postures is beneficial. Hence, an instructor is required to monitor the correctness of Yoga postures. However, at times, it is difficult to have an instructor. This study aims to provide a system that will act as a personal Yoga instructor and practitioners can practice Yoga in their comfort zone. The device is interactive and provides audio guidance to perform different Yoga asanas. It makes the use of a camera to capture the picture of the person performing Yoga in a particular position. This captured pose is compared with the benchmark postures. A pretrained deep learning model is used for the classification of different Yoga postures using a standard dataset. Based on the comparison, the practitioner's posture will be corrected using a voice message to move the body parts in a certain direction. As the device performs all the operations in real-time, it has a quick response time of a few seconds. Currently, this work aids the practitioners in performing five Asanas, namely, Ardha Chandrasana /Half-moon pose, Tadasana /Mountain pose, Trikonasana /Triangular pose, Veerabhadrasana /Warrior pose, and Vrikshasana /Tree pose.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Smart Yoga Instructor for Guiding and Correcting Yoga Postures in Real Time

Kishore

Bindu²,

Manjunath

2022

International Journal of Yoga

View full text Add to dashboard Cite

show abstract

“…In its simplest form, such an analysis amounts to detecting whether the subject performs the exercise correctly or not. This was achieved several works [1,6,5] by exploiting 2D poses extracted from the input images. In particular, Dittakavi et al [5] detected which joints need to be fixed by finding the overall joint angle distribution of the dataset and detecting poses in which a joint angle is an anomaly.…”

Section: Physical Exercise Analysismentioning

confidence: 99%

“…This was achieved several works [1,6,5] by exploiting 2D poses extracted from the input images. In particular, Dittakavi et al [5] detected which joints need to be fixed by finding the overall joint angle distribution of the dataset and detecting poses in which a joint angle is an anomaly. This framework operates on single frames, as opposed to our method which operates on entire sequences.…”

Section: Physical Exercise Analysismentioning

confidence: 99%

“…While useful, such classification-based approaches offer limited information to the user, as they do not provide them with any feedback about the specific type of mistakes they made. Moreover, most of existing works operate on 2D pose inputs [1,2,3,6,5]. Similar to [4], we also design our framework to work with 3D poses enabling us to be robust to ambiguities found in 2D poses.…”

Section: Physical Exercise Analysismentioning

confidence: 99%

“…There is therefore a growing need for computer-aided exercise feedback strategies. A few recent works have addressed this problem [1,2,3,4,5,6]. However, they focus only on identifying whether an exercise is performed correctly or not [1,6], or they rely on hard-coded rules based on joint angles that cannot easily be extended to new exercises [2,3,4].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

3D Pose Based Feedback for Physical Exercises

Zhao¹,

Kiciroglu²,

Hugues³

et al. 2022

Preprint

View full text Add to dashboard Cite

Unsupervised self-rehabilitation exercises and physical training can cause serious injuries if performed incorrectly. We introduce a learning-based framework that identifies the mistakes made by a user and proposes corrective measures for easier and safer individual training. Our framework does not rely on hard-coded, heuristic rules. Instead, it learns them from data, which facilitates its adaptation to specific user needs. To this end, we use a Graph Convolutional Network (GCN) architecture acting on the user's pose sequence to model the relationship between the the body joints trajectories. To evaluate our approach, we introduce a dataset with 3 different physical exercises. Our approach yields 90.9% mistake identification accuracy and successfully corrects 94.2% of the mistakes.

show abstract

PoseScript: 3D Human Poses from Natural Language

Ginger

Weinzaepfel²,

Lucas³

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Automatically producing instructions to modify one's posture could open the door to endless applications, such as personalized coaching and in-home physical therapy. Tackling the reverse problem (i.e., refining a 3D pose based on some natural language feedback) could help for assisted 3D character animation or robot teaching, for instance. Although a few recent works explore the connections between natural language and 3D human pose, none focus on describing 3D body pose differences. In this paper, we tackle the problem of correcting 3D human poses with natural language. To this end, we introduce the Pose-Fix dataset, which consists of several thousand paired 3D poses and their corresponding text feedback, that describe how the source pose needs to be modified to obtain the target pose. We demonstrate the potential of this dataset on two tasks: (1) text-based pose editing, that aims at generating corrected 3D body poses given a query pose and a text modifier; and (2) correctional text generation, where instructions are generated based on the differences between two body poses. The dataset and the code are available at https://europe.naverlabs.com/research/ computer-vision/posefix/.

show abstract

Pose Tutor: An Explainable System for Pose Correction in the Wild

Cited by 35 publications

References 23 publications

Smart Yoga Instructor for Guiding and Correcting Yoga Postures in Real Time

Smart Yoga Instructor for Guiding and Correcting Yoga Postures in Real Time

3D Pose Based Feedback for Physical Exercises

PoseScript: 3D Human Poses from Natural Language

Contact Info

Product

Resources

About