MovingFashion: a Benchmark for the Video-to-Shop Challenge

Godi, Marco; Joppi, Christian; Skenderi, Geri; Cristani, Marco

doi:10.1109/wacv51458.2022.00059

Cited by 12 publications

(6 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This makes still image animation ambiguous, reducing the informativeness of the video. An alternative Dataset Instance # image/video # class Stanford Online Products [99] Furniture 59,551 11,318 DeepFashion [83] Clothing 239,557 33,881 DeepFashion2 [38] Clothing 224,114 45,417 Street2Shop [42] Clothing >420,357 204,795 MovingFashion [39] Clothing ∼ 15,045 15,045 Table 1. Publicly Available Products Datasets for Evaluating VM Representations solution that can enable the post-production step to enhance the central route of persuasion is to extract selling points [41] from the VMs or descriptions by summarization or image/video captioning techniques, then explicitly present this information as voice-over or captions in the generated videos.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

AI-Empowered Persuasive Video Generation: A Survey

Liu¹,

Han²

2021

Preprint

View full text Add to dashboard Cite

Promotional videos are rapidly becoming a popular medium for persuading people to change their behaviours in many settings (e.g., online shopping, social enterprise initiatives). Today, such videos are often produced by professionals, which is a time-, labour-and cost-intensive undertaking. In order to produce such contents to support a large applications (e.g., e-commerce), the field of artificial intelligence (AI)-empowered persuasive video generation (AIPVG) has gained traction in recent years. This field is interdisciplinary in nature, which makes it challenging for new researchers to grasp. Currently, there is no comprehensive survey of AIPVG available. In this paper, we bridge this gap by reviewing key AI techniques that can be utilized to automatically generate persuasive videos. We offer a first-of-its-kind taxonomy which divides AIPVG into three major steps: 1) visual material understanding, which extracts information from the visual materials (VMs) relevant to the target of promotion; 2) visual storyline generation, which shortlists and arranges high-quality VMs into a sequence in order to compose a storyline with persuasive power; and 3) post-production, which involves background music generation and still image animation to enhance viewing experience. We also introduce the evaluation metrics and datasets commonly adopted in the field of AIPVG. We analyze the advantages and disadvantages of the existing works belonging to the above-mentioned steps, and discuss interesting potential future research directions.

show abstract

Section: Discussionmentioning

confidence: 99%

“…All of them are E-commerce platforms selling clothing. MovingFashion [39] is a publicly available video dataset for clothing items. Each video is associated with a distinct image of a shop, and is retrieved from Net-A-Porter 7 , Instagram or Tik Tok.…”

Section: Evaluating Vm Representationsmentioning

confidence: 99%

AI-Empowered Persuasive Video Generation: A Survey

Liu¹,

Han²

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…When a subject approaches the totem, a small video sequence is captured by the system and the crop of the face is used to identify the person through the reidentification module. We then use the pretrained model introduced in MovingFashion [20] to perform the detection of the top and bottom part of the outfit. The crops of the clothes are fed to a simple convolutional network that extracts a collection of attributes, derived from the DeepFashion dataset [27], which we then use in a naive manner to infer the style of the person's clothing.…”

Section: Body Clothing Analysismentioning

confidence: 99%

I-MALL An Effective Framework for Personalized Visits. Improving the Customer Experience in Stores

Becattini

Becchi

Ferracani

et al. 2022

Proceedings of the 1st Workshop on Multimedia Computing Towards Fashion Recommendation

View full text Add to dashboard Cite

show abstract

“…Instance # image/video # class Stanford Online Products [69] Furniture 59,551 11,318 DeepFashion [151] Clothing 239,557 33,881 DeepFashion2 [152] Clothing 224,114 45,417 Street2Shop [153] Clothing >420,357 204,795 MovingFashion [154] Clothing ⇠ 15,045 15,045…”

Section: Datasetmentioning

confidence: 99%

“…MovingFashion [154] is a publicly available video dataset for clothing items. Each video is associated with a distinct image of a shop, and is retrieved from Net-A-Porter 7 , Instagram or Tik Tok.…”

Section: Datasetsmentioning

confidence: 99%

AI-empowered promotional video generation

Liu¹

View full text Add to dashboard Cite

Professor Han Yu for his guidance and kind support throughout my Ph.D. study. This thesis would not have been possible without your insightful advice and fruitful discussion. I would also thank Prof Chunyan Miao, Prof Boyang Li, and Dr. Zhiqi Shen for their support and constructive conversations. It is a great honor for me to pursuit my Ph.D. under your guidance. I wish to thank my collaborators who worked on promotional video generation together and provided support along my Ph.D. journey:

show abstract

MovingFashion: a Benchmark for the Video-to-Shop Challenge

Cited by 12 publications

References 10 publications

AI-Empowered Persuasive Video Generation: A Survey

AI-Empowered Persuasive Video Generation: A Survey

I-MALL An Effective Framework for Personalized Visits. Improving the Customer Experience in Stores

AI-empowered promotional video generation

Contact Info

Product

Resources

About