A Richly Annotated Pedestrian Dataset for Person Retrieval in Real Surveillance Scenarios

Li, Dangwei; Zhang, Zhang; Chen, Xiaotang; Huang, Kaiqi

doi:10.1109/tip.2018.2878349

Cited by 130 publications

(90 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…DenseNet-169 training is accomplished using full body images for gender classification. AVSS 2018 challenge II [5], RAP [17], PETA [18], and DukeMTMC-reID [21] datasets are used to collect 1,45,386 full body images. Each image is resized to 350 × 140 resolution to preserve the spatial ratio of full body.…”

Section: Densenet-169 Training For Gender Classificationmentioning

confidence: 99%

“…The proposed algorithm achieves highest average IoU for 20 sequences {TS. 2,3,5,7,8,9,13,14,16,17,20,22,24,26,27,34,36,37, 38, 39} out of 41. The above sequences cover all the difficulty levels in the dataset.…”

Section: Performance Analysismentioning

confidence: 99%

“…Pose annotations in the dataset are avoided by PGDM which transfer the existing pose estimation model knowledge to pedestrian attribute database. The evaluation uses RAP [17], PETA [18] and PA-100K [19] large scale pedestrian attribute datasets. Given the input image, Part-based Convolutional Baseline (PCB) [13] 1501 [20] dataset.…”

Section: Introductionmentioning

confidence: 99%

“…Researchers are extensively using PETA [18], Market-1501 [20], DukeMTMC-reID [21], CUHK03 [22] and RAP [17] large scale pedestrian datasets for person identification or re-identification. These datasets consist of a gallery of cropped images of persons and annotated soft biometric attributes.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Visual appearance based person retrieval in unconstrained environment videos

Galiyawala

Raval

Dave

2019

Image and Vision Computing

View full text Add to dashboard Cite

Visual appearance-based person retrieval is a challenging problem in surveillance. It uses attributes like height, cloth color, cloth type and gender to describe a human. Such attributes are known as soft biometrics. This paper proposes person retrieval from surveillance video using height, torso cloth type, torso cloth color and gender. The approach introduces an adaptive torso patch extraction and bounding box regression to improve the retrieval. The algorithm uses fine-tuned Mask R-CNN and DenseNet-169 for person detection and attribute classification respectively. The performance is analyzed on AVSS 2018 challenge II dataset and it achieves 11.35% improvement over state-of-the-art based on average Intersection over Union measure.

show abstract

Section: Densenet-169 Training For Gender Classificationmentioning

confidence: 99%

Section: Performance Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Visual appearance based person retrieval in unconstrained environment videos

Galiyawala

Raval

Dave

2019

Image and Vision Computing

View full text Add to dashboard Cite

show abstract

“…We use RAP-2.0 [32] as our benchmark pedestrian attribute recognition dataset. This dataset contains 84,928 images which were divided into three parts, of which 50,957 for training, 16,986 for validation, and 16,985 for testing.…”

Section: Datasetmentioning

confidence: 99%

Pedestrian Attribute Recognition with Graph Convolutional Network in Surveillance Scenarios

2019

View full text Add to dashboard Cite

Pedestrian attribute recognition is to predict a set of attribute labels of the pedestrian from surveillance scenarios, which is a very challenging task for computer vision due to poor image quality, continual appearance variations, as well as diverse spatial distribution of imbalanced attributes. It is desirable to model the label dependencies between different attributes to improve the recognition performance as each pedestrian normally possesses many attributes. In this paper, we treat pedestrian attribute recognition as multi-label classification and propose a novel model based on the graph convolutional network (GCN). The model is mainly divided into two parts, we first use convolutional neural network (CNN) to extract pedestrian feature, which is a normal operation processing image in deep learning, then we transfer attribute labels to word embedding and construct a correlation matrix between labels to help GCN propagate information between nodes. This paper applies the object classifiers learned by GCN to the image representation extracted by CNN to enable the model to have the ability to be end-to-end trainable. Experiments on pedestrian attribute recognition dataset show that the approach obviously outperforms other existing state-of-the-art methods.

show abstract