Fashionpedia: Ontology, Segmentation, and an Attribute Localization Dataset

Jia, Menglin; Shi, Mengyun; Sirotenko, Mikhail; Cui, Yin; Cardie, Claire; Hariharan, Bharath; Adam, Hartwig; Belongie, Serge

doi:10.1007/978-3-030-58452-8_19

Cited by 79 publications

(62 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For readers who are interested in knowing more about the performance of different state-of-theart model architectures, we refer readers to Fashionpedia projects where researchers created a large-scale fashion dataset fully annotated by fashion experts and then compared the accuracy of the current state-of-the-art models. 70 Segmentation techniques. The results show that the quality of the segmentations generated by our A.I.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

The exploration of artificial intelligence application in fashion trend forecasting

Shi

Chussid

Yang

et al. 2021

Textile Research Journal

Self Cite

View full text Add to dashboard Cite

Fashion trends today are changing much faster than ever before. Timely and reliable trend forecasting is, therefore, critical in the fashion industry. Traditional fashion forecasting requires professionals to abstract image-based information across design collections and time intervals from around the world, which is extremely time-consuming and labor intensive. Considering the financial cost associated with manual labeling and the accuracy of classifications based upon human subjective judgment, this explorative study proposes a data-driven quantitative abstracting approach using an artificial intelligence (A.I.) algorithm. Firstly, an A.I. model was trained to be familiar with fashion images from a large-scale dataset under different scenarios such as online stores and street snapshots; secondly, the model could detect garments and classify clothing attributes such as fabric textures, garment style, and design details from runway photos and videos; thirdly, the model could summarize fashion trends from the attributes it developed. The adoption of an A.I. algorithm proved to be an objective and systematic computerized method of interpreting fashion dynamics in a more efficient, accurate, sustainable, and cost-effective way.

show abstract

Section: Discussionmentioning

confidence: 99%

“…Owing to the lack of human-annotated ground truth labels for the larger sample of runway show images, this study is more focused on the qualitative analysis of generated attributes. In future studies, a more quantitative analysis should be conducted on runway show images, as demonstrated in Jia et al 70…”

Section: Limitations and Future Workmentioning

confidence: 99%

The exploration of artificial intelligence application in fashion trend forecasting

Shi

Chussid

Yang

et al. 2021

Textile Research Journal

Self Cite

View full text Add to dashboard Cite

show abstract

“…Perhaps, the most similar dataset in this regard is the recent Fashionpedia [24], a dataset providing attributes and localizations of 27 apparel categories. However, the dataset is proposed for the fashion domain which limits its utility for general purpose object recognition task.…”

Section: Rival10mentioning

confidence: 99%

A Comprehensive Study of Image Classification Model Sensitivity to Foregrounds, Backgrounds, and Visual Attributes

Moayeri¹,

Pope²,

Balaji³

et al. 2022

Preprint

View full text Add to dashboard Cite

While datasets with single-label supervision have propelled rapid advances in image classification, additional annotations are necessary in order to quantitatively assess how models make predictions. To this end, for a subset of ImageNet samples, we collect segmentation masks for the entire object and 18 informative attributes. We call this dataset RIVAL10 (RIch Visual Attributes with Localization), consisting of roughly 26k instances over 10 classes. Using RIVAL10, we evaluate the sensitivity of a broad set of models to noise corruptions in foregrounds, backgrounds and attributes. In our analysis, we consider diverse state-of-the-art architectures (ResNets, Transformers) and training procedures (CLIP, SimCLR, DeiT, Adversarial Training). We find that, somewhat surprisingly, in ResNets, adversarial training makes models more sensitive to the background compared to foreground than standard training. Similarly, contrastively-trained models also have lower relative foreground sensitivity in both transformers and ResNets. Lastly, we observe intriguing adaptive abilities of transformers to increase relative foreground sensitivity as corruption level increases. Using saliency methods, we automatically discover spurious features that drive the background sensitivity of models and assess alignment of saliency maps with foregrounds. Finally, we quantitatively study the attribution problem for neural features by comparing feature saliency with ground-truth localization of semantic attributes.

show abstract

“…In addition, there are a number of domain specific (i.e. "finegrained") datasets covering object categories such as airplanes [44,59], birds [60,3,57,36], dogs [32,51,42], fashion [31], flowers [47,48], food [4,28], leaves [39], vehicles [37,41,62,17], and, of course, human faces [29,50,22,5]. Most closely related to our work are the existing iNaturalist species classification datasets [58,2], which contain a set of coarse and fine-grained species classification problems.…”

Section: Fine-grained Datasetsmentioning

confidence: 99%

Benchmarking Representation Learning for Natural World Image Collections

Horn

Cole

Beery

et al. 2021

2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

Self Cite

100

View full text Add to dashboard Cite

Recent progress in self-supervised learning has resulted in models that are capable of extracting rich representations from image collections without requiring any explicit label supervision. However, to date the vast majority of these approaches have restricted themselves to training on standard benchmark datasets such as ImageNet. We argue that fine-grained visual categorization problems, such as plant and animal species classification, provide an informative testbed for self-supervised learning. In order to facilitate progress in this area we present two new natural world visual classification datasets, iNat2021 and NeWT. The former consists of 2.7M images from 10k different species uploaded by users of the citizen science application iNaturalist. We designed the latter, NeWT, in collaboration with domain experts with the aim of benchmarking the performance of representation learning algorithms on a suite of challenging natural world binary classification tasks that go beyond standard species classification. These two new datasets allow us to explore questions related to large-scale representation and transfer learning in the context of finegrained categories. We provide a comprehensive analysis of feature extractors trained with and without supervision on ImageNet and iNat2021, shedding light on the strengths and weaknesses of different learned features across a diverse set of tasks. We find that features produced by standard supervised methods still outperform those produced by self-supervised approaches such as SimCLR. However, improved self-supervised learning methods are constantly being released and the iNat2021 and NeWT datasets are a valuable resource for tracking their progress.

show abstract

Fashionpedia: Ontology, Segmentation, and an Attribute Localization Dataset

Cited by 79 publications

References 42 publications

The exploration of artificial intelligence application in fashion trend forecasting

The exploration of artificial intelligence application in fashion trend forecasting

A Comprehensive Study of Image Classification Model Sensitivity to Foregrounds, Backgrounds, and Visual Attributes

Benchmarking Representation Learning for Natural World Image Collections

Contact Info

Product

Resources

About