Person Re-identification by Descriptive and Discriminative Classification

By Martin Hirzer, Csaba Beleznai, Peter M. Roth, Horst Bischof. Scandinavian Conference on Image Analysis, 2011

1,033 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/16c7c31a7553d99f1837fc6e88e77b5ccbb346b8

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

Citations in the corpus

• Multi-Task Learning with Low Rank Attribute Embedding for Person Re-Identification (2015, methodology): “4.3.2 PRID Following the protocol in [16], we use images of 100 persons from camera A as the probe set, and 649 persons in camera B as the gallery set, excluding all training samples.”
• A Memorizing and Generalizing Framework for Lifelong Person Re-Identification (2023, methodology): “LReID-Unseen Subset: We merge the testing sets of 7 popular person ReID datasets: VIPeR [83], PRID [84], GRID [85], i-LIDS [86], CUHK01 [87], CUHK02 [88], and SenseReID [89] to form the unseen testing subset, named as “LReID-Unseen”.”
• Scalable Person Re-identification on Supervised Smoothed Manifold (2017, background): “Most of them require human feedback in-theloop [3, 17], such as Post-rank OPtimisation (POP) [29], Human Verification Incremental Learning (HVIL) [51].”
• Person Re-identification by Multi-Channel Parts-Based CNN with Improved Triplet Loss Function (2016, background): “Given a query person’s image, in order to find the correct matches among a large set of candidate images captured by different cameras, two crucial problems must be addressed.”
• Deep Learning for Person Re-Identification: A Survey and Outlook (2020, methodology): “MSMT17 [35]) and 7 video datasets (PRID-2011 [126], iLIDSVID [6], MARS [7], Duke-Video [128], Duke-Tracklet [177],”
• Lifelong Person Re-Identification via Adaptive Knowledge Accumulation (2021, methodology): “Specifically, we first merge VIPeR [6], PRID [8], GRID [25], i-LIDS [41], CUHK01 [17], CUHK02 [16], SenseReID [50] in accordance with their original train/test splits as a new benchmark.”
• Human Semantic Parsing for Person Re-identification (2018, methodology): “In addition to these datasets that were used in evaluation, we utilize 3DPeS [4], CUHK01 [23], CUHK02 [22], PRID [17], PSDB [41], Shinpuhkan [20] and VIPeR [15] datasets to augment our training data.”
• Deep Transfer Learning for Person Re-Identification (2016, background): “Five widely used datasets are used including two large datasets (CUHK03 [19] and Market1501 [20]) and three small ones (VIPeR [21], PRID [27] and CUHK01 [28]).”
• Quality Aware Network for Set to Set Recognition (2017, methodology): “Results of evaluation obeying “10-fold cross validation” on PRID2011 and iLIDS-VID are shown in Table 1 and Table 2.”
• Large scale metric learning from equivalence constraints (2012, background): “KISSME 20% 62% 81% 92% SDALF [5] 20% 50% 70% 85% DDC [10] 19% 52% 69% 80% PRDC [23] 16% 54% 76% 87%”
• A survey of approaches and trends in person re-identification (2014, neutral): “A fewothermulti-cameradatasets like Chokepoint [101], Terrascope [102], Person Re-ID dataset (PRID) [76], SAIVT-SoftBio [103] and CUHK02 [75] should be mentioned in this context as they can be very well applied to evaluation of person Re-ID methods.”
• Person Recognition System Based on a Combination of Body Images from Visible Light and Thermal Cameras (2017, background): “Although there are several public databases for person identification research such as the CUHK01 [49] and CUHK03 databases [50], the VIPeR database [51], the iLIDS-VID database [52], and the PRID2011 database [53], these databases cannot be used in our research because they contain only visible lig”
• Jointly Attentive Spatial-Temporal Pooling Networks for Video-Based Person Re-identification (2017, methodology): “We evaluate our model for video-based person re-id on three different datasets: iLIDS-VID [27], PRID-2011 [8] and MARS [39].”
• A Survey of Deep Active Learning (2020, methodology): “[4, 230] PRID [93], MARS [256], BDD100K [246], DukeMTMC-VideoReID [237], CityPersons [249], Caltech Pedestrian[52] Person Re-id”
• Re-id: Hunting Attributes in the Wild (2014, methodology): “20 40 60 80 100 0 0.2 0.4 0.6 0.8 1 % M a tc h Rank VIP SDALF ELF BR−SVM KISS FUSIA eSDC 20 40 60 80 100 0 0.2 0.4 0.6 0.8 1 % M a tc h Rank GRID SDALF ELF BR−SVM KISS FUSIA 20 40 60 80 100 0 0.2 0.4 0.6 0.8 1 % M a tc h Rank PRID SDALF ELF BR−SVM KISS FUSIA 20 40 60 80 100 0 0.2 0.4 0.6 0.8 1 % M a”
• 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”
• Local Fisher Discriminant Analysis for Pedestrian Re-identification (2013, methodology): “Model estimation using Adaboost to improve ranking results has also been proposed [18].”
• Person Re-identification: Past, Present and Future (2016, methodology): “It is generally believed that iLIDS-VID is more challenging than PRID-2011 due to extremely heavy occlusion.”
• Patch-based Knowledge Distillation for Lifelong Person Re-Identification (2022, background): “It consists of twelve ReID datasets, including five seen datasets (Market-1501 [73], CUHK-SYSU [64], DukeMTMC-reID [48], MSMT17_V2 [61] and CUHK03 [32]) for non-forgetting evaluation and seven unseen datasets (VIPeR [16], PRID [22], GRID [39], iLIDS [74], CUHK01 [31], CUHK02 [30] and SenseReID [70])”
• Brownian descriptor: A rich meta-feature for appearance matching (2014, methodology): “The proposed descriptor is referred to as Brownian descriptor due to its analogy to the Brownian covariance.”