SPDA-CNN: Unifying Semantic Part Detection and Abstraction for Fine-Grained Recognition

By Han Zhang, Tao Xu, Mohamed Elhoseiny, Xiaolei Huang, Shaoting Zhang, Ahmed M. Elgammal, Dimitris N. Metaxas. 2016

297 citations recorded by the source corpus.

Subfields: 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 Computer Vision and Pattern Recognition

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

Citations in the corpus

• Fine-Grained Image Classification via Combining Vision and Language (2017, neutral): “50%) or even part annotations [41, 46] (85.”
• A Generative Adversarial Approach for Zero-Shot Learning from Noisy Texts (2017, methodology): “The region proposals by [49] are passed through ROI pooling layer and fed into an n-way classifier (n is the number of semantic parts plus background) and a bounding box regressor.”
• Weakly Supervised Learning of Part Selection Model with Spatial Constraints for Fine-Grained Image Classification (2017, methodology): “Second, inspired by (Zhang et al. 2016b), we will focus on the work of feature encoding for further improvement.”
• Multi-branch and Multi-scale Attention Learning for Fine-Grained Visual Categorization (2020, background): “3 SPDA-CNN [5] X 85.”
• Learning a Discriminative Filter Bank Within a CNN for Fine-Grained Recognition (2016, background): “As long as the layer is properly initialized, the whole network can be trained in an end-to-end fashion just once, which is more efficient compared with the multistage training strategy of previous works [29, 50, 18].”
• Learning Multi-attention Convolutional Neural Network for Fine-Grained Image Recognition (2017, methodology): “To extract both local and global features from an image, we follow previous works [1, 18, 33] to take as input for PartCNN from both part-level patches and object-level images.”
• Adaptive Triplet Model for Fine-Grained Visual Categorization (2018, background): “Remarkably, ourmodels are comparable to or even better than thosemethods [4], [13]–[15], [20], [21], [27], [39] with extra annotations, including bounding box and part annotation.”
• Hierarchical Bilinear Pooling for Fine-Grained Visual Recognition (2018, methodology): “The proposed HBP( relu 5 3 + relu 5 2 + relu 5 1) achieves better result compared with PN-CNN [2], SPDA-CNN [35] and B-CNN [17] even without bbox and part annotation, which demonstrates the effectiveness of our model.”
• Look Closer to See Better: Recurrent Attention Convolutional Neural Network for Fine-Grained Image Recognition (2017, methodology): “• SPDA-CNN [32]: semantic part detection and abstraction proposes to generate part candidates and extract features by detection/classification networks.”
• Audio Visual Attribute Discovery for Fine-Grained Object Recognition (2018, methodology): “…by the use of deep convolutional neural network, the performance is heavily dependent on the supervised labels to construct the discriminative feature representation, e.g. precise parts (Xie et al. 2013; Zhang et al. 2016a; 2014) or textual annotations (Reed et al. 2016; Vedaldi et al. 2014).”
• Object-Part Attention Model for Fine-Grained Image Classification (2017, methodology): “Moreover, our approach outperforms methods that use both object and part annotations [6], [23].”
• Bilinear CNNs for Fine-grained Visual Recognition (2015, methodology): “For example, the SPDA-CNN [69] trains better part detectors and feature representations using the VGG-D network and report 84.”
• WildFish: A Large Benchmark for Fish Recognition in the Wild (2018, background): “Furthermore, the extra annotations of bounding boxes and parts are often used as a strong supervision to detect distinct regions of fine-grained categories [21, 25, 46, 47].”
• Attention Convolutional Binary Neural Tree for Fine-Grained Visual Categorization (2019, methodology): “1 SPDA-CNN [50] CaffeNet 85.”
• Fine-Grained Representation Learning and Recognition by Exploiting Hierarchical Semantic Embedding (2018, methodology): “We compare our HSE framework with 17 state-of-the-art methods, including Deep Localization, Alignment and Classi cation (DeepLAC) [24], Semantic Part Detection and Abstraction (SPDA-CNN) [46], Part-RCNN [47], Part Alignment-based (PA-CNN) [20], Pose Normalized CNN (PNCNN) [1], Picking Deep Filter Re”