Automated Flower Classification over a Large Number of Classes

By Maria-Elena Nilsback, Andrew Zisserman. 2008

4,982 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology, Plant Science

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

In Media Technology

In Plant Science

Citations in the corpus

An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (2020, methodology): “We transfer the models trained on these dataset to several benchmark tasks: ImageNet on the original validation labels and the cleaned-up ReaL labels (Beyer et al., 2020), CIFAR-10/100 (Krizhevsky, 2009), Oxford-IIIT Pets (Parkhi et al., 2012), and Oxford Flowers-102 (Nilsback & Zisserman, 2008).”
Do Better ImageNet Models Transfer Better? (2018, neutral): “[19] 20 5,011/4,952 11-point mAP Describable Textures (DTD) [10] 47 3,760/1,880 top-1 Oxford-IIIT Pets [53] 37 3,680/3,369 mean per-class Caltech-101 [20] 102 3,060/6,084 mean per-class Oxford 102 Flowers [52] 102 2,040/6,149 mean per-class”
On Robustness and Transferability of Convolutional Neural Networks (2020, neutral): “CIFAR-100 [38], DTD [9], Flowers102 [47], Pets [49], Sun397 [68], SVHN [46].”
Big Transfer (BiT): General Visual Representation Learning (2019, methodology): “These tasks include ImageNet’s ILSVRC-2012 [10], CIFAR-10/100 [27], Oxford-IIIT Pet [41], Oxford Flowers-102 [39] (including few-shot variants), and the 1000-sample VTAB-1k benchmark [66], which consists of 19 diverse datasets.”
Bootstrap Your Own Latent: A New Approach to Self-Supervised Learning (2020, methodology): “We perform transfer via linear classification and fine-tuning on the same set of datasets as in [8], namely Food-101 dataset [89], CIFAR-10 [78] and CIFAR-100 [78], Birdsnap [90], the SUN397 scene dataset [79], Stanford Cars [91], FGVC Aircraft [92], the PASCAL VOC 2007 classification task [80], the”
Learning Visual Features from Large Weakly Supervised Data (2015, methodology): “…a range of computer-vision tasks: (1) the MIT Indoor dataset [38], (2) the MIT SUN dataset [55], (3) the Stanford 40 Actions dataset [57], (4) the Oxford Flowers dataset [33], (5) the Sports dataset [17], (6) the ImageNet ILSVRC 2014 dataset [42], and (7) the Pascal VOC 2007 dataset [11].”
Predicting Deep Zero-Shot Convolutional Neural Networks Using Textual Descriptions (2015, background): “The Oxford Flower-102 dataset[19] contains 102 classes with a total of 8189 images.”
Emerging Properties in Self-Supervised Vision Transformers (2021, methodology): “We use ImageNet-1k [60] (“Inet”), Places205 [84], PASCAL VOC [24] and Oxford-102 flowers (“FLOWERS”) [46].”
An Empirical Study of Training Self-Supervised Vision Transformers (2021, methodology): “We study four datasets used in [15]: CIFAR-10/100 [25], Oxford Flowers-102 [32], and Oxford-IIIT-Pets [34].”
Training data-efficient image transformers & distillation through attention (2020, neutral): “ImageNet [39] 1,281,167 50,000 1000 iNaturalist 2018 [24] 437,513 24,426 8,142 iNaturalist 2019 [25] 265,240 3,003 1,010 Flowers-102 [36] 2,040 6,149 102 Stanford Cars [28] 8,144 8,041 196 CIFAR-100 [29] 50,000 10,000 100 CIFAR-10 [29] 50,000 10,000 10”
Grafit: Learning fine-grained image representations with coarse labels (2020, methodology): “Flowers-102 [41] 2,040 6,149 102 Stanford Cars [35] 8,144 8,041 196 Food101 [7] 75,750 25,250 101”
Projected GANs Converge Faster (2021, background): “org), Oxford Flowers (1360 images) [42], photographs of landscapes (4319 images; flickr.”
CLIPDraw: Exploring Text-to-Drawing Synthesis through Language-Image Encoders (2021, methodology): “Commonly used datasets include Oxford-120 Flowers (Nilsback & Zisserman, 2008), CUB-200 Birds (Wah et al., 2011), and COCO (Lin et al., 2014), all of which are comprised of natural images along with a set of cap-tions describing them.”