From generic to specific deep representations for visual recognition

Azizpour, Hossein; Razavian, Ali Sharif; Sullivan, Josephine; Maki, Atsuto; Carlsson, Stefan

doi:10.1109/cvprw.2015.7301270

Cited by 344 publications

(321 citation statements)

References 33 publications

Supporting

Mentioning

304

Contrasting

Order By: Relevance

“…The model saved at 200,000 iterations was always the source model. This is in line with the research done by Azizpour et al [21], which found that early stopping was less beneficial than overfitting, although the benefit of overfitting diminishes beyond 200,000 iterations.…”

Section: Experimental Framework and Datasetsupporting

confidence: 79%

See 1 more Smart Citation

Improving optimization of convolutional neural networks through parameter fine-tuning

Becherer

Pecarina

Nykl

et al. 2017

Neural Comput & Applic

View full text Add to dashboard Cite

In recent years, convolutional neural networks have achieved state-of-the-art performance in a number of computer vision problems such as image classification. Prior research has shown that a transfer learning technique known as parameter finetuning wherein a network is pre-trained on a different dataset can boost the performance of these networks. However, the topic of identifying the best source dataset and learning strategy for a given target domain is largely unexplored. Thus, this research presents and evaluates various transfer learning methods for fine-grained image classification as well as the effect on ensemble networks. The results clearly demonstrate the effectiveness of parameter fine-tuning over random initialization. We find that training should not be reduced after transferring weights, larger, more similar networks tend to be the best source task, and parameter fine-tuning can often outperform randomly initialized ensembles. The experimental framework and findings will help to train models with improved accuracy.

show abstract

Section: Experimental Framework and Datasetsupporting

confidence: 79%

“…The work of both Razavian et al [20] and Azizpour et al [21] also focuses on applying CNNs to other problems. In general, they find that CNNs coupled with SVMs provide competitive results to existing state-of-the-art solutions for many datasets.…”

Section: Related Work In Transfer Learning For Cnnsmentioning

confidence: 99%

Improving optimization of convolutional neural networks through parameter fine-tuning

Becherer

Pecarina

Nykl

et al. 2017

Neural Comput & Applic

View full text Add to dashboard Cite

show abstract

“…For instance, reusing the first layers of a network have been shown to be an extremely good base representation of the visual information [17,29]. More specifically, applications of pre-trained CNN to the problem of visual instance retrieval have been studied in [6,8,30] on the classic Oxford5k, Paris5k and Holidays benchmarks. Building on these analysis, we use the VGG16 CNN architecture [34] as our base network (see Fig.…”

Section: Cnn Methodsmentioning

confidence: 99%

Visual Link Retrieval in a Database of Paintings

Séguin

Striolo

diLenardo

et al. 2016

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. This paper examines how far state-of-the-art machine vision algorithms can be used to retrieve common visual patterns shared by series of paintings. The research of such visual patterns, central to Art History Research, is challenging because of the diversity of similarity criteria that could relevantly demonstrate genealogical links. We design a methodology and a tool to annotate efficiently clusters of similar paintings and test various algorithms in a retrieval task. We show that pretrained convolutional neural network can perform better for this task than other machine vision methods aimed at photograph analysis. We also show that retrieval performance can be significantly improved by fine-tuning a network specifically for this task.

show abstract

“…Using CNNs trained for object recognition has a long history in computer vision and machine learning. While they have been known to yield good results on supervised image classification tasks such as MNIST for a long time [17], recently they were not only shown to outperform classical methods in large scale image classification tasks [13], object detection [9] and semantic segmentation [8] but also to produce features that transfer between tasks [7], [2]. This recent success story has been made possible through optimized implementations for high-performance computing systems, as well as the availability of large amounts of labeled image data through, e.g., the ImageNet dataset [19].…”

Section: Related Workmentioning

confidence: 99%

Multimodal deep learning for robust RGB-D object recognition

Eitel

Springenberg

Spinello

et al. 2015

2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

525

401

View full text Add to dashboard Cite

Abstract-Robust object recognition is a crucial ingredient of many, if not all, real-world robotics applications. This paper leverages recent progress on Convolutional Neural Networks (CNNs) and proposes a novel RGB-D architecture for object recognition. Our architecture is composed of two separate CNN processing streams -one for each modality -which are consecutively combined with a late fusion network. We focus on learning with imperfect sensor data, a typical problem in real-world robotics tasks. For accurate learning, we introduce a multi-stage training methodology and two crucial ingredients for handling depth data with CNNs. The first, an effective encoding of depth information for CNNs that enables learning without the need for large depth datasets. The second, a data augmentation scheme for robust learning with depth images by corrupting them with realistic noise patterns. We present stateof-the-art results on the RGB-D object dataset [15] and show recognition in challenging RGB-D real-world noisy settings.

show abstract

From generic to specific deep representations for visual recognition

Cited by 344 publications

References 33 publications

Improving optimization of convolutional neural networks through parameter fine-tuning

Improving optimization of convolutional neural networks through parameter fine-tuning

Visual Link Retrieval in a Database of Paintings

Multimodal deep learning for robust RGB-D object recognition

Contact Info

Product

Resources

About