Multiclass recognition and part localization with humans in the loop

By Catherine Wah, Steve Branson, Pietro Perona, Serge J. Belongie. 2011

200 citations recorded by the source corpus.

Subfields: Artificial Intelligence, 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 Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop (2015, methodology): “Clever methods have been proposed to improve the efﬁciency of human-in-the-loop annotation (e.g., [2, 5, 13, 20, 21]).”
The Unreasonable Effectiveness of Noisy Data for Fine-Grained Recognition (2015, methodology): “ction. Considerable work has also gone into the challenging task of curating negrained datasets [4,58,27,30,31,59,65,60,70] and developing interactive methods for recognition with a human in the loop [6,62,61,63]. While these works have demonstrated eective strategies for collecting images of ne-gra”
Learning to Compare: Relation Network for Few-Shot Learning (2017, background): “CUB (Caltech-UCSD Birds-200-2011) [39] contains 11,788 images of 200 bird species.”
Evaluation of output embeddings for fine-grained image classification (2014, methodology): “We evaluate SJE on three datasets: Caltech UCSD Birds (CUB) [54] and Stanford Dogs (Dogs)(2) [26] are finegrained, and Animals With Attributes (AWA) [29] is a standard attribute dataset for zero-shot classification.”
Bird Species Categorization Using Pose Normalized Deep Convolutional Nets (2014, methodology): “Among this large body of work, it is a goal of our paper to empirically investigate which methods and techniques are most important toward achieving good performance.”
Label-Embedding for Image Classification (2015, neutral): “…adds (1) an expanded related work section; (2) a detailed description of the learning procedure for ALE; (3) additional comparisons with random embed-dings [14] and embeddings derived automatically from textual corpora [40], [20]; (4) additional zero-short learning experiments, which show…”
Leveraging the Wisdom of the Crowd for Fine-Grained Recognition (2016, unknown): “Fine-grained recognition is challenging.”
Label-Embedding for Attribute-Based Classification (2013, unknown): “Finally, we draw conclusions.”
A codebook-free and annotation-free approach for fine-grained image categorization (2012, methodology): “Fine-grained image classification [1] [13] [3] [31] [24] [10] [27] refers to the task of classifying objects that belong to the same basic-level category.”
The Ignorant Led by the Blind: A Hybrid Human–Machine Vision System for Fine-Grained Categorization (2014, methodology): “The computer vision algorithms described in this section obtain state-of-the-art performance on CUB-200-2011 (Wah et al. 2011) without any interactive component.”
Best of both worlds: Human-machine collaboration for object annotation (2015, methodology): “In computer vision with human-in-the-loop approaches, human intervention has ranged from binary question-and-answer [6, 59, 60] to attribute-based feedback [40, 39, 34] to free-form object annotation [58].”
Learning Features and Parts for Fine-Grained Recognition (2014, background): “Another line of research focuses on putting humans in the loop [33], [33]–[36].”
Unsupervised Domain Adaptation for Zero-Shot Learning (2015, background): “The results are slightly better than that without adaptation on AwA; but on the more challenging CUB dataset, their adaptations have an adverse effect.”
Local Alignments for Fine-Grained Categorization (2014, methodology): “In (Yao et al. 2012) the same authors propose to randomly generate thousands of templates, which after being convolved with the unseen images lead to very high-dimensional representations.”
Learning a Deep Embedding Model for Zero-Shot Learning (2016, methodology): “Extensive experiments carried out on four benchmarks including AwA [21], CUB [44] and large scale ILSVRC 2010 and ILSVRC 2012 [5] show that our model beats all the stateof-the-art models presented to date, often by a large margin.”