Nonparametric Part Transfer for Fine-Grained Recognition

By Christoph Göring, Erik Rodner, Alexander Freytag, Joachim Denzler. 2014

122 citations recorded by the source corpus.

Subfields: Aerospace Engineering, 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 Aerospace Engineering

1. Erik Rodner
2. Marcel Simon
3. Trevor Darrell
4. Joachim Denzler
5. Ross B. Girshick
6. Björn Barz
7. Qi Tian
8. Pietro Perona

In Computer Vision and Pattern Recognition

1. Dacheng Tao
2. Luc Van Gool
3. Trevor Darrell
4. Yi Yang
5. Xiaogang Wang
6. Ming-Hsuan Yang
7. Bernt Schiele
8. Alan L. Yuille

Citations in the corpus

• Part-Based R-CNNs for Fine-Grained Category Detection (2014, background): “[22] transfers part annotations from objects with similar global shape as non-parametric part detections.”
• The Unreasonable Effectiveness of Noisy Data for Fine-Grained Recognition (2015, background): “The majority of research in fine-grained recognition has focused on developing improved models for classification [2, 5, 7, 15, 17, 18, 24, 30, 37, 44, 45, 58, 59, 60, 63, 64].”
• Neural Activation Constellations: Unsupervised Part Model Discovery with Convolutional Networks (2015, methodology): “There are many supervised methods for part model learning which rely on ground-truth part or bounding box annotations [39, 18, 29].”
• The Devil is in the Tails: Fine-grained Classification in the Wild (2017, background): “Most fine-grained research papers present a novel model for classification [4, 6, 7, 12, 15, 17, 25, 31, 40, 45, 46, 57, 58, 59, 61, 62].”
• Part Detector Discovery in Deep Convolutional Neural Networks (2014, positive): “Our method adapts the approach of [26] replacing the SIFT and color name features by the activations of a deep neural network and the nonparametric part transfer by the presented part detection approach.”
• SPDA-CNN: Unifying Semantic Part Detection and Abstraction for Fine-Grained Recognition (2016, methodology): “[16] extracted handcrafted features from each part for the final classification; this method achieved promising classification results on the Caltech-UCSD birds datasets [37, 38], because all 15 small semantic parts of the bird were used.”
• Local Novelty Detection in Multi-class Recognition Problems (2015, methodology): “Furthermore, local learning principles have been used in fine-grained recognition to learn exemplar-specific feature representations [12] and for nonparametric part transfer [13].”
• Deep LAC: Deep localization, alignment and classification for fine-grained recognition (2015, unknown): “[9] 57.”
• Weakly Supervised Fine-Grained Image Categorization (2015, background): “Some works [10, 13] transfer the part annotations from objects in training images to those sharing similar shapes in testing images instead of applying object / part detectors.”
• Coarse-to-Fine Description for Fine-Grained Visual Categorization (2016, methodology): “Various methods have been proposed for part localization [2]–[7], [10]–[13], [15], [24], [32]–[35].”
• Deep Learning on Small Datasets without Pre-Training using Cosine Loss (2019, methodology): “8% on CUB by making use of the object part annotations provided for the training set [13].”
• Fine-grained recognition without part annotations (2015, unknown): “8 Nonparametric [21] BBox+Parts BBox 57.”