Learning Discriminative Appearance-Based Models Using Partial Least Squares
By William Robson Schwartz, Larry S. Davis. 2009 XXII Brazilian Symposium on Computer Graphics and Image Processing, 2009
391 citations recorded by the source corpus.
Subfields: Computer Vision and Pattern Recognition
View Paper: https://www.semanticscholar.org/paper/89dbfb9b75d3902748d73bfb5965e7d11e83c10e
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
Citations in the corpus
- Person Re-identification: What Features Are Important? (2012, methodology): “In contrast to [12] that requires labelled gallery images to discover gallery-specific feature importance, our method is fully unsupervised.”
- Sparse Coding and Dictionary Learning for Symmetric Positive Definite Matrices: A Kernel Approach (2012, positive): “We used the modified ETHZ dataset [31].”
- Person re-identification by symmetry-driven accumulation of local features (2010, methodology): “Similarly, in [19], a high-dimensional signature composed by texture, gradient and color information is projected into a low-dimensional discriminant latent space by Partial Least Squares (PLS) reduction.”
- MARS: A Video Benchmark for Large-Scale Person Re-Identification (2016, methodology): “3DPES [3] and ETH [32] are of similar scales.”
- Deep Metric Learning for Person Re-identification (2014, methodology): “Five popular databases were built for person reidentification: VIPeR [4], i-LIDS [20], ETHZ [21], CIVAR [22] and CUHK Campus [5].”
- Deep feature learning with relative distance comparison for person re-identification (2015, neutral): “Adaboost [2], Xing's [15], LMNN [16], ITML [17], PLS [28], Bhat.”
- 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”
- A survey of approaches and trends in person re-identification (2014, positive): “Partial least squares (PLS) technique is employed to not only find discriminative weighting for color, texture and edge features but also as a means to reduce descriptor dimensionality in [39].”
- Unsupervised Salience Learning for Person Re-identification (2013, methodology): “To qualitatively compare with sophiscated supervised learning methods, Figure 4(a) shows the feature weighting map estimated by partial least square (PLS) [26].”
- PersonNet: Person Re-identification with Deep Convolutional Neural Networks (2016, methodology): “Schwartz and Davis propose a discriminative appearance based model using partial least squares, in which multiple visual features: texture, gradient and color features are combined [31].”
- Relaxed Pairwise Learned Metric for Person Re-identification (2012, methodology): “, [1– 10]) in recent years, and also various benchmark datasets [11, 8, 6] have been published.”
- Learning to rank in person re-identification with metric ensembles (2015, methodology): “Schwartz and Davis propose a discriminative appearance based model using partial least squares, in which multiple visual features: texture, gradient and color features are combined [35].”
- BiCov: a novel image representation for person re-identification and face verification (2012, methodology): “The proposed representation has been experimentally validated on two datasets for person re-identification (VIPeR [10] and ETHZ [6, 20]) and one for face verification (LFW [12]).”
- Person Re-Identification by Iterative Re-Weighted Sparse Ranking (2015, methodology): “3 PLS [12] 79 – – 74.”
- Local Descriptors Encoded by Fisher Vectors for Person Re-identification (2012, methodology): “…re-identification are: color (widely used since the color of clothing constitutes simple but efficient visual signatures) [1], HOG like signatures [2, 3], texture [4–6], differential filters [6], Haar-like fil-ters [7], co-occurrence matrices [3], interest points [8], e.g. SURF and SIFT [9, 10] or the…”
- Clothing Attributes Assisted Person Reidentification (2015, methodology): “In person reidentification, VIPeR [13], i-LIDS [4], and ETHZ [15] are the most frequently used datasets.”