Fine-Grained Visual Classification of Aircraft

Maji, Subhransu; Rahtu, Esa; Kannala, Juho; Blaschko, Matthew; Vedaldi, Andrea

doi:10.48550/arxiv.1306.5151

Cited by 283 publications

(413 citation statements)

References 3 publications

Supporting

Mentioning

409

Contrasting

Unclassified

Order By: Relevance

“…Meta-Test Datasets We apply our model trained from source database to 6 benchmark datasets such as 1) CIFAR-10 ( Krizhevsky et al, 2009), 2) CIFAR-100 (Krizhevsky et al, 2009), 3) MNIST (Le-Cun & Cortes, 2010), 4) SVHN (Netzer et al, 2011), 5) Aircraft (Maji et al, 2013), and 6) Oxford-IIIT Pets (Parkhi et al, 2012). On CIFAR10 and CIFAR100, the generator generates 500 neural architectures and we select 30 architectures based on accuracies predicted by the predictor.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Rapid Neural Architecture Search by Learning to Generate Graphs from Datasets

Lee¹,

Hyung²,

Hwang³

2021

Preprint

View full text Add to dashboard Cite

Despite the success of recent Neural Architecture Search (NAS) methods on various tasks which have shown to output networks that largely outperform humandesigned networks, conventional NAS methods have mostly tackled the optimization of searching for the network architecture for a single task (dataset), which does not generalize well across multiple tasks (datasets). Moreover, since such task-specific methods search for a neural architecture from scratch for every given task, they incur a large computational cost, which is problematic when the time and monetary budget are limited. In this paper, we propose an efficient NAS framework that is trained once on a database consisting of datasets and pretrained networks and can rapidly search for a neural architecture for a novel dataset. The proposed MetaD2A (Meta Dataset-to-Architecture) model can stochastically generate graphs (architectures) from a given set (dataset) via a cross-modal latent space learned with amortized meta-learning. Moreover, we also propose a meta-performance predictor to estimate and select the best architecture without direct training on target datasets. The experimental results demonstrate that our model meta-learned on subsets of ImageNet-1K and architectures from NAS-Bench 201 search space successfully generalizes to multiple unseen datasets including CIFAR-10 and CIFAR-100, with an average search time of 33 GPU seconds. Even under MobileNetV3 search space, MetaD2A is 5.5K times faster than NSGANetV2, a transferable NAS method, with comparable performance. We believe that the MetaD2A proposes a new research direction for rapid NAS as well as ways to utilize the knowledge from rich databases of datasets and architectures accumulated over the past years. Code is available at https://github.com/HayeonLee/MetaD2A.

show abstract

Section: Methodsmentioning

confidence: 99%

“…The number of classes is 10 denoting from digit 1 to 10 and the number of training/test images is 73257/26032, respectively. 5) Aircraft (Maji et al, 2013) This is fine-grained classification benchmark dataset containing 10K images from 30 different aircraft classes. We resize all images into 32×32.…”

Section: Search Spacementioning

confidence: 99%

Rapid Neural Architecture Search by Learning to Generate Graphs from Datasets

Lee¹,

Hyung²,

Hwang³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…For example, the class bird is further split into multiple categories based on species. More information can be seen for Caltech-UCSD Birds 200 (CUB), Cars and aircraft in [52], [22] and [33] respectively. The Table 3 shows the performance comparison for few of the specific image augmentation technqiues designed for fine grained image recognition.…”

Section: Fine-grained Image Recognitionmentioning

confidence: 99%

Survey: Image Mixing and Deleting for Data Augmentation

Naveed¹

2021

Preprint

View full text Add to dashboard Cite

Data augmentation has been widely used to improve deep nerual networks performance. Numerous approaches are suggested, for example, dropout, regularization and image augmentation, to avoid over-ftting and enhancing generalization of neural networks. One of the sub-area within data augmentation is image mixing and deleting. This specific type of augmentation either mixes two images or delete image regions to hide or make certain characteristics of images confusing for the network to force it to emphasize on overall structure of object in image. The model trained with this approach has shown to perform and generalize well as compared to one trained without imgage mixing or deleting. Additional benefit achieved with this method of training is robustness against image corruptions. Due to its low compute cost and success in recent past, many techniques of image mixing and deleting are proposed. This paper provides detailed review on these devised approaches, dividing augmentation strategies in three main categories cut and delete, cut and mix and mixup. The second part of paper emprically evaluates these approaches for image classification, finegrained image recognition and object detection where it is shown that this category of data augmentation improves the overall performance for deep neural networks.

show abstract

“…Aircraft [43] The dataset consists of 10,000 aircraft images in 100 classes, with 66/34 or 67/33 training/testing images per class. Birdsnap [44] The dataset has 49,829 images of 500 species of North American Birds.…”

Section: A1 Image Datasets Descriptionmentioning

confidence: 99%

Frustratingly Easy Transferability Estimation

Huang¹,

Wei²,

Rong³

et al. 2021

Preprint

View full text Add to dashboard Cite

Transferability estimation has been an essential tool in selecting a pre-trained model and the layers of it to transfer, so as to maximize the performance on a target task and prevent negative transfer. Existing estimation algorithms either require intensive training on target tasks or have difficulties in evaluating the transferability between layers. We propose a simple, efficient, and effective transferability measure named TransRate. With single pass through the target data, TransRate measures the transferability as the mutual information between the features of target examples extracted by a pre-trained model and labels of them. We overcome the challenge of efficient mutual information estimation by resorting to coding rate that serves as an effective alternative to entropy. TransRate is theoretically analyzed to be closely related to the performance after transfer learning. Despite its extraordinary simplicity in 10 lines of codes, TransRate performs remarkably well in extensive evaluations on 22 pre-trained models and 16 downstream tasks.

show abstract

Fine-Grained Visual Classification of Aircraft

Cited by 283 publications

References 3 publications

Rapid Neural Architecture Search by Learning to Generate Graphs from Datasets

Rapid Neural Architecture Search by Learning to Generate Graphs from Datasets

Survey: Image Mixing and Deleting for Data Augmentation

Frustratingly Easy Transferability Estimation

Contact Info

Product

Resources

About