An Evaluation Of Design Choices For Pedestrian Attribute Recognition In Video

Specker, Andreas; Schumann, Arne; Beyerer, Jürgen

doi:10.1109/icip40778.2020.9191264

Cited by 9 publications

(6 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most of the work in literature [10,11,12] relies on PAR as the basis for attribute-based person retrieval. Given an pedestrian image X, such approaches predict an attribute vector y ∈ {0, 1} M by computing a probability vector p ∈ [0, 1] M and applying a threshold.…”

Section: Methodsmentioning

confidence: 99%

“…Please note that the semantics of attributes gets lost during inference since the matching is done in the learned, abstract feature space, similar to conventional image-based person re-identification. Other works build on attribute classifiers to deal with the task of attribute-based person retrieval [10,11,12]. For example, best practices presented in [10] indicate that separate classifiers for unrelated attributes and attention-mechanisms can improve the performance.…”

Section: Related Workmentioning

confidence: 99%

“…Again, the Euclidean distance is leveraged for the person retrieval. The same applies to [12] which takes a look into several approaches for PAR in videos, also evaluating the suitability of the methods for the task of attributebased person retrieval.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Improving Attribute-Based Person Retrieval By Using A Calibrated, Weighted, And Distribution-Based Distance Metric

Specker

Beyerer

2021

2021 IEEE International Conference on Image Processing (ICIP)

Self Cite

View full text Add to dashboard Cite

Typically, person re-identification systems use so-called query images of a person to find occurrences of a person in surveillance footage. In real-world scenarios, however, often only witness descriptions and no images of a person-ofinterest are available. In such cases, attribute-based retrieval can be performed based on pedestrian attribute recognition approaches. Current methods rely on calculating the Euclidean distance between query attributes and attribute predictions of gallery samples to compute ranking result lists. However, these methods do not consider the output distributions of the attribute classifier. We propose to do so by introducing a distance computation that remedies the effects of unbalanced distributions of attribute predictions. Moreover, we adapt a calibration technique to reduce the negative influences further. We also propose to weight the attributes during retrieval based on their prediction errors. In total, our approach surpasses the retrieval performance achieved by the Euclidean distance by a large margin on several datasets. In figures, we were able to increase the mAP on the RAP-2.0 and Market-1501 datasets by 4.2% and 2.4% points, respectively.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Improving Attribute-Based Person Retrieval By Using A Calibrated, Weighted, And Distribution-Based Distance Metric

Specker

Beyerer

2021

2021 IEEE International Conference on Image Processing (ICIP)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Security services already started to rely on unmanned aerial vehicles (UAV) for short-term use cases such as recon flights at disaster scenarios, which opens up new possibilities, as initial work has already demonstrated their suitability for the detection and localization of persons and objects [15][16][17]. The use of facial recognition for tracking or recognition in classical video surveillance setups has also experienced significant improvement in the past [18][19][20] as well as attribute-based person re-identification [21,22]. Even the field of action recognition has also seen a variety of methods emerge in the near past, promising ever better accuracies and real-time processing, which is essential for serious applications [23,24].…”

Section: Smart Video Surveillance Using Digital Skeletonsmentioning

confidence: 99%

“…The first step of a general solution approach is the extraction of the posture of several human bodies simultaneously from this image material. This step is also called Multi-Person Pose Estimation [20,21] or just Human Pose Estimation [22,23]. The outputs are keypoints of the digital skeleton (see Figure 2), which relate to various body parts as joints or some relevant body parts like eyes or nose.…”

Section: Technical Principles Of the Smart Subsystemmentioning

confidence: 99%

Perception of Risks and Usefulness of Smart Video Surveillance Systems

2022

View full text Add to dashboard Cite

The number of video cameras in public places increases due to different reasons such as detecting dangers (e.g., thefts, robberies, terrorist attacks) and security breaches in crowds. The application of video surveillance systems is sometimes evaluated ambivalently; therefore, the presented study focuses on factors influencing the acceptance of a privacy-friendly, smart video surveillance system. Overall, 216 persons aged between 18 and 81 years participated in an online survey. In terms of the perceived usefulness, there are significant interactions of public spaces × gender and public spaces × time of day. In addition, the assessment of different privacy levels of a video surveillance system differ significantly in terms of perceived risk. Interestingly, men rate the risk concerning their own privacy significantly higher than women do. Participants rate the presented system as fairly useful and slightly risky for their own privacy. The findings of the presented exploratory study provide insight into how people perceive smart video surveillance. These findings have the potential to support the conditions of the use of smart video surveillance systems and to address the possibly affected individuals. smart video surveillance; intelligent video surveillance; machine learning; human pose estimation; privacy protection; online survey; exploratory study; time of day; public areas; gender; perceived privacy risk

show abstract