Silhouette-Based Action Recognition Using Simple Shape Descriptors

Gościewska, Katarzyna; Frejlichowski, Dariusz

doi:10.1007/978-3-030-00692-1_36

Cited by 7 publications

(5 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Yang et al 8 calculated the relative position between joints to represent the action spatial information, and employs the Fourier time pyramid to model the temporal dynamics. There are also some silhouette-based 45,46 methods for action recognition. Yi githan Dedeo glu et al 45 classified actions using object silhouettes.…”

Section: Related Workmentioning

confidence: 99%

“…They employed an adaptive background subtraction model for target segmentation and a template matching‐based supervised learning method for object classification. Katarzyna Gościewska et al 46 used a single scalar shape measure to represent each silhouette in an action sequence, and then combined the scalars into a vector that represents the entire sequence. However, hand‐crafted features are not enough to effectively characterize spatial–temporal information, and it is difficult to adapt to complex scene changes.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

CHAN: Skeleton based action recognition by multi‐level feature learning

Lü

Gong

Zhou

et al. 2023

Computer Animation & Virtual

View full text Add to dashboard Cite

Skeleton‐based action recognition has been continuously and intensively studied. However, dynamic 3D skeleton data are difficult to be popularized in practical applications due to the restricted data acquisition conditions. Although the action recognition method based on 2D pose information extracted from RGB video can effectively avoid the influence of complex background, it is susceptible to factors such as video jitter and joint overlap. To reduce the interference of the aforementioned factors, we use two‐dimensional skeletal joint coordinate modal information to represent the changes in human body posture. First, we use a target detector and pose estimation algorithm to obtain the joint coordinates of each frame sample from RGB video. Then the feature extraction network is combined to perform multi‐level feature learning to establish correspondence between actions and corresponding multi‐level features. Finally, the hierarchical attention mechanism is introduced to design the model named CHAN. By calculating the association between elements, the weight of the action classification is redistributed. Extensive experiments on three datasets demonstrate the effectiveness of our proposed method.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

CHAN: Skeleton based action recognition by multi‐level feature learning

Lü

Gong

Zhou

et al. 2023

Computer Animation & Virtual

View full text Add to dashboard Cite

show abstract

“…In addition, [16] proposed a long-term motion descriptor called sequential Deep Trajectory Descriptor (sDTD) which feeds a CNN-RNN network with dense trajectories to learn an effective representation for long-term motion. On the other hand, silhouettes were exploited by [17], [18] after extracting them using background subtraction. Automatic feature extraction from RGB images through deep learning was also suggested in many works.…”

Section: Vision-based Human Activity Recognitionmentioning

confidence: 99%

Vision-Based Multi-Modal Framework for Action Recognition

Romaissa

Oussalah

Nini

2021

2020 25th International Conference on Pattern Recognition (ICPR)

View full text Add to dashboard Cite

Human activity recognition plays a central role in the development of intelligent systems for video surveillance, public security, health care and home monitoring, where detection and recognition of activities can improve the quality of life and security of humans. Typically, automated, intuitive and real-time systems are required to recognize human activities and identify accurately unusual behaviors in order to prevent dangerous situations. In this work, we explore the combination of three modalities (RGB, depth and skeleton data) to design a robust multi-modal framework for vision-based human activity recognition. Especially, spatial information, body shape/posture and temporal evolution of actions are highlighted using illustrative representations obtained from a combination of dynamic RGB images, dynamic depth images and skeleton data representations. Therefore, each video is represented with three images that summarize the ongoing action. Our framework takes advantage of transfer learning from pre-trained models to extract significant features from these newly created images. Next, we fuse extracted features using Canonical Correlation Analysis and train a Long Short-Term Memory network to classify actions from visual descriptive images. Experimental results demonstrated the reliability of our feature-fusion framework that allows us to capture highly significant features and enables us to achieve the state-of-the-art performance on the public UTD-MHAD and NTU RGB+D datasets.

show abstract

“…Quan et al [35] leveraged multiple view information and pair it with silhouettes to learn binary features to represent the shape. Silhouette shape and optical point descriptors were also utilized to recognize human actions [17] [14]. Similarly, a sequence of images with silhouettes was used to conduct action recognition [5].…”

Section: Related Workmentioning

confidence: 99%

Single Image Shape-from-Silhouettes

Wang

2020

Proceedings of the 28th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Recovering a 3D shape representation from one single image input has been attempted in recent years. Most of the works obtain 3D models from multiple images at different perspectives or ground truth CAD models. However, multiple images from different perspectives or 3D CAD models are not always available in real applications. In this work, we present a novel shape-from-silhouette method based on just a single image, which is an end-to-end learning framework relying on view synthesis and shape-from-silhouette methodology to reconstruct a 3D shape. The reconstructed 3D mesh can approach the real shape of target objects by constraining the silhouettes from both horizontal and vertical directions, especially for those objects with occlusions. Our proposed method achieves stateof-the-art performance on the ShapeNet dataset compared with other recent approaches targeting 3D reconstruction from a single image. Without requiring labor-intensive and time-consuming human annotations, the work has a broad potential to be applied in real-world applications. CCS CONCEPTS • Information systems → Multimedia content creation.

show abstract

Silhouette-Based Action Recognition Using Simple Shape Descriptors

Cited by 7 publications

References 22 publications

CHAN: Skeleton based action recognition by multi‐level feature learning

CHAN: Skeleton based action recognition by multi‐level feature learning

Vision-Based Multi-Modal Framework for Action Recognition

Single Image Shape-from-Silhouettes

Contact Info

Product

Resources

About