Custom Pictorial Structures for Re-identification

By Dong Seon Cheng, Marco Cristani, Michele Stoppa, Loris Bazzani, Vittorio Murino. 2011

722 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Yi Yang
2. Liang Zheng
3. Xiaogang Wang
4. Qi Tian
5. Luc Van Gool
6. Dacheng Tao
7. Ming-Hsuan Yang
8. Tao A. Xiang

Citations in the corpus

• Faceless Person Recognition: Privacy Implications in Social Media (2016, background): “Pose-independent descriptors have been explored for the body region [28, 38, 39, 29, 30].”
• Person Recognition in Personal Photo Collections (2015, background): “Similarly, pose-independent descriptors can be built for the body [8, 17, 40].”
• A Pose-Sensitive Embedding for Person Re-identification with Expanded Cross Neighborhood Re-ranking (2017, methodology): “A more fine-grained pose representation based on Pictorial Structures was first used in [5] to focus on matching between individual body parts.”
• RGB-Infrared Cross-Modality Person Re-identification (2017, neutral): “) Datasets ID# images# cameras# RGB IR VIPER [7] 632 1,264 2 yes no iLIDS [54] 119 476 2 yes no CAVIAR [5] 72 610 2 yes no PRID2011 [11] 200 971 2 yes no CUHK01 [19] 972 1,942 2 yes no SYSU [8] 502 24,448 2 yes no CUHK03 [20] 1467 13,164 6 yes no Market [53] 1501 32,668 6 yes no MARS [52] 1261 1,191”
• Cross-View Asymmetric Metric Learning for Unsupervised Person Re-Identification (2017, background): “Model SDALF CPS UDML GL GTS SDC CAMEL [8] [7] [24] [12] [29] [35]”
• Re-identification with RGB-D Sensors (2012, background): “Most of the re-identification approaches build on appearance-based features [1, 11, 3] and this prevents from focusing on re-id scenarios where the clothing may change.”
• Recurrent Convolutional Network for Video-Based Person Re-identification (2016, background): “The first of these employs invariant feature based methods that attempt to extract features that are both discriminative and invariant to environmental and view-point changes [23, 3, 6].”
• Pose-Invariant Embedding for Deep Person Re-Identification (2017, methodology): “The closest works to PoseBox are [8, 7], which construct the pictorial structure (PS), a similar concept to PoseBox.”
• Partial Person Re-Identification (2015, methodology): “The gap become even bigger on P-iLIDS which contains more naturally occluded person images than P-CAVIAR.”
• A Siamese Long Short-Term Memory Architecture for Human Re-identification (2016, background): “Most of the works on human re-identification concentrates on either developing a new feature representation [9, 27, 33, 36, 58, 59, 63] or learning a new distance”
• Pedestrian Alignment Network for Large-scale Person Re-Identification (2017, methodology): “The PoseBox is a strengthened version of the Pictorial Structures proposed in [Cheng et al., 2011].”
• Beyond Part Models: Person Retrieval with Refined Part Pooling (2017, methodology): “[4] employ pictorial structure to parse the pedestrian into semantic parts.”
• SVDNet for Pedestrian Retrieval (2017, background): “[30] propose the PoseBox structure similar to the pictorial structure [6] to learn pose invariant embeddings.”
• Person Re-identification by Multi-Channel Parts-Based CNN with Improved Triplet Loss Function (2016, methodology): “Once the feature extraction is completed, these methods usually choose a standard distance measure such as l1-norm based distance (L1norm) [40], Bhattacharyya distance (Bhat)[14], or Mahalanobis distance (Maha)[33] to determine the similarity between pairs.”