Deep filter banks for texture recognition and segmentation

By Mircea Cimpoi, Subhransu Maji, Andrea Vedaldi. 2015

671 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

Citations in the corpus

Do Better ImageNet Models Transfer Better? (2018, methodology): “Several papers have proposed methods to make better use of CNN features and thus improve the efficacy of transfer learning [43, 11, 42, 23, 73, 63, 13, 41, 54].”
Spatial Transformer Networks (2015, background): “…instance of the digit, reducing the complexity of the task for the classification network, and does not over fit on simpler data e.g. R. Interestingly, the transformation of inputs for all ST models leads to a “standard” upright posed digit – this is the mean pose found in the training data.”
Unsupervised Learning from Narrated Instruction Videos (2015, background): “A similar representation (aggregated into compact VLAD descriptor) was shown to work well recently for a variety of recognition tasks [8].”
Convolutional Two-Stream Network Fusion for Video Action Recognition (2016, background): “Using deeper models, such as the very deep networks in [23] can, however, lead to even better performance in image recognition tasks [5, 15, 27].”
Context Encoding for Semantic Segmentation (2018, methodology): “Prior work adopts dense CRF taking FCN outputs to refine the segmentation boundaries [5, 7], and CRFRNN achieves end-to-end learning of CRF with FCN [58].”
Tree Species Identification from Bark Images Using Convolutional Neural Networks (2018, methodology): “segmentation algorithms, [23] used an SVM to classify texture materials, notably on the Flickr Material Dataset [24].”
Deep Structure-Revealed Network for Texture Recognition (2020, methodology): “Method F V-VGGVD [3] CVPR2015 VGGVD FC-VGGVD [3] CVPR2015 VGGVD B-CNN [13] CVPR2016 VGGVD B-CNN [27] CVPR2016 Resnet18 LFV [24] ICCV2017 VGGVD FASON[5] CVPR2017 VGGVD DeepTEN [33] CVPR2017 Resnet50 DeepTEN [27] CVPR2017 Resnet18 DEP[27] CVPR2018 Resnet50 DEP[27] CVPR2018 Resnet18 MAP-net [31] ICCV20”
Picking Deep Filter Responses for Fine-Grained Image Recognition (2016, background): “[6] pool features from the convolutional layers, and achieve considerable improvements for texture recognition.”
A Comprehensive Look at Coding Techniques on Riemannian Manifolds (2018, neutral): “classification [6], [10].”
Fine-Grained Image Analysis With Deep Learning: A Survey (2021, neutral): “compared with fully-connected outputs [135], [136], [137].”
Aggregating Local Deep Features for Image Retrieval (2015, methodology): “Recently, research attention shifted from the features extracted from the fully-connected layers to the features from the deep convolutional layers of CNNs [5, 22, 14] (below we refer to these features as deep convolutional features).”
Cross-Dimensional Weighting for Aggregated Deep Convolutional Features (2015, methodology): “[8] also recently proposed using Fisher Vector aggregation of convolutional features for texture recognition.”
Beyond Correlation Filters: Learning Continuous Convolution Operators for Visual Tracking (2016, background): “Recently, the last (deep) convolutional layers were shown to be more beneficial for image classification [8,33].”
Semantic-Aware Scene Recognition (2019, methodology): “Cimpoi et al. [25] used material and texture information extracted from a CNN-ﬁlter bank and from a bag-of-visual-words to increase generalization for enhanced scene recognition.”
Aggregating Local Deep Features for Image Retrieval (2015, methodology): “Recently, research attention shifted from the features extracted from the fully-connected layers to the features from the deep convolutional layers of CNNs [5, 22, 14] (below we refer to these features as deep convolutional features).”
Training with Confusion for Fine-Grained Visual Classification (2017, methodology): “Robust image representations such as CNN filter banks [4], VLAD [14] and Fisher vectors [37] have been prior techniques at tackling fine-grained classification.”
AGA: Attribute-Guided Augmentation (2016, methodology): “, has been successfully used for object detection [14, 27], scene classification [15, 8], texture classification [6] and domain adaptation [9], using various transfer mechanisms.”