Deep LAC: Deep localization, alignment and classification for fine-grained recognition

By Di Lin, Xiaoyong Shen, Cewu Lu, Jiaya Jia. 2015

360 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

1. Dacheng Tao
2. Jiashi Feng
3. Luc Van Gool
4. Trevor Darrell
5. Chunhua Shen
6. Alan L. Yuille
7. Yu Qiao
8. Qi Tian

In Computer Vision and Pattern Recognition

1. Dacheng Tao
2. Luc Van Gool
3. Yi Yang
4. Ming-Hsuan Yang
5. Xiaogang Wang
6. Trevor Darrell
7. Chunhua Shen
8. Jiashi Feng

Citations in the corpus

• Recent advances in convolutional neural networks (2015, background): “[107] incorporate part localization, alignment, and classification into one recognition system which is called Deep LAC.”
• Do Semantic Parts Emerge in Convolutional Neural Networks? (2016, background): “2015) and fine-grained recognition (Hariharan et al. 2015; Lin et al. 2015; Zhang et al. 2014).”
• Coarse2Fine: a two-stage training method for fine-grained visual classification (2019, methodology): “[14] proposed a framework, Deep-LAC in which the alignment and classification errors are propagated back to the localization layers using their proposed valve linkage function (VLF).”
• Fine-Grained Image Analysis With Deep Learning: A Survey (2021, background): “6% – Deep LAC [67] CVPR 2015 BBox+Parts BBox Alex-Net 227× 227 80.”
• A survey on deep learning-based fine-grained object classification and semantic segmentation (2017, methodology): “The deep LAC [28] incorporates part localization, alignment, and classification in one deep neural network.”
• Part-Stacked CNN for Fine-Grained Visual Categorization (2015, methodology): “The complete PS-CNN model with a bounding-box and 15 object parts achieves 76% accuracy, which is comparable with part-based R-CNN [51], while being slightly lower than several most recent state-of-the-art methods [22, 21, 35] due to the effectiveness-efficiency tradeoff.”
• Coarse-to-Fine Description for Fine-Grained Visual Categorization (2016, methodology): “In order to release the dependencies on part annotations during the testing stage, Lin et al. [13], Branson et al. [3], Liu et al. [6], [7], and Zhang et al. [11] generate local parts by transferring the part annotations from training images to testing images.”
• Mask-CNN: Localizing parts and selecting descriptors for fine-grained bird species categorization (2018, methodology): “, [7, 8, 10, 11], directly used the deep convolutional descriptors and encoded them into a single representation, without evaluating the usefulness of the obtained object/part deep descriptors.”
• Multi-Attention Multi-Class Constraint for Fine-grained Image Recognition (2018, background): “Recently, with the popularity of DNNs, the tasks of object part localization and feature representation can be both learned in a more effective way [7,8,9,10,11].”