To Click or Not To Click

Song, Yang; Реди, Мириам; Vallmitjana, Jordi; Jaimes, Alejandro

doi:10.1145/2983323.2983349

Cited by 70 publications

(41 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the context of online advertising, researchers have found that image quality properties can impact the user experience of the ad viewed [72]. Aesthetically appealing preview thumbnails increase the clickthorough probability of a video [63]. In recent work, Schifanella et al showed how existing features for aesthetics, embedded in topic-specific aesthetic models, can be used to surface beautiful but hard-to-find pictures and that content quality is only weakly correlated with its popularity [60].…”

Section: Related Workmentioning

confidence: 99%

Beautiful and Damned. Combined Effect of Content Quality and Social Ties on User Engagement

Aiello

Schifanella

Реди

et al. 2017

IEEE Trans. Knowl. Data Eng.

Self Cite

View full text Add to dashboard Cite

User participation in online communities is driven by the intertwinement of the social network structure with the crowd-generated content that flows along its links. These aspects are rarely explored jointly and at scale. By looking at how users generate and access pictures of varying beauty on Flickr, we investigate how the production of quality impacts the dynamics of online social systems. We develop a deep learning computer vision model to score images according to their aesthetic value and we validate its output through crowdsourcing. By applying it to over 15B Flickr photos, we study for the first time how image beauty is distributed over a large-scale social system. Beautiful images are evenly distributed in the network, although only a small core of people get social recognition for them. To study the impact of exposure to quality on user engagement, we set up matching experiments aimed at detecting causality from observational data. Exposure to beauty is double-edged: following people who produce high-quality content increases one's probability of uploading better photos; however, an excessive imbalance between the quality generated by a user and the user's neighbors leads to a decline in engagement. Our analysis has practical implications for improving link recommender systems.

show abstract

Section: Related Workmentioning

confidence: 99%

Beautiful and Damned. Combined Effect of Content Quality and Social Ties on User Engagement

Aiello

Schifanella

Реди

et al. 2017

IEEE Trans. Knowl. Data Eng.

Self Cite

View full text Add to dashboard Cite

show abstract

“…More recent methods focused on a few characteristics that were identified as the most important ones for thumbnail selection, and relate to the representativeness and the aesthetic quality or attractiveness of the visual content. These methods are based either on traditional feature extraction and clustering algorithms [23,25], or on the use of deep network architectures [9]. Finally, a few recent works associate video thumbnail selection with the users' intentions when searching for video content online, and propose multimodal approaches for dynamic video thumbnail selection, according to textual user queries [14,17,26,30].…”

Section: Introductionmentioning

confidence: 99%

Combining Adversarial and Reinforcement Learning for Video Thumbnail Selection

Apostolidis

Adamantidou

Mezaris

et al. 2021

Proceedings of the 2021 International Conference on Multimedia Retrieval

View full text Add to dashboard Cite

This paper presents a new method for unsupervised video thumbnail selection. The developed network architecture selects video thumbnails based on two criteria: the representativeness and the aesthetic quality of their visual content. Training relies on a combination of adversarial and reinforcement learning. The former is used to train a discriminator, whose goal is to distinguish the original from a reconstructed version of the video based on a small set of candidate thumbnails. The discriminator's feedback is a measure of the representativeness of the selected thumbnails. This measure is combined with estimates about the aesthetic quality of the thumbnails (made using a SoA Fully Convolutional Network) to form a reward and train the thumbnail selector via reinforcement learning. Experiments on two datasets (OVP and Youtube) show the competitiveness of the proposed method against other SoA approaches. An ablation study with respect to the adopted thumbnail selection criteria documents the importance of considering the aesthetics, and the contribution of this information when used in combination with measures about the representativeness of the visual content. CCS CONCEPTS• Computing methodologies → Video summarization; Machine learning algorithms.

show abstract

“…A majority of the existing video summarization methods follow a common recipe: step 1) video segmentation, step 2) segment ranking and step 3) segment selection [3,6,12,21]. Methods vary in how segmentation is performed and how individual segments are ranked.…”

Section: Introductionmentioning

confidence: 99%

“…This method relies upon the availability of suitable training data. Early video summarization methods were unsupervised [3,12,21]; however, with the recent availability of high-quality video summarization datasets, many newer methods are supervised [6,24].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Real-time Video Summarization on Commodity Hardware

Taylor

Qureshi

2018

Proceedings of the 12th International Conference on Distributed Smart Cameras

View full text Add to dashboard Cite

We present a method for creating video summaries in real-time on commodity hardware. Real-time here refers to the fact that the time required for video summarization is less than the duration of the input video. First, low-level features are use to discard undesirable frames. Next, video is divided into segments, and segment-level features are extracted for each segment. Tree-based models trained on widely available video summarization and computational aesthetics datasets are then used to rank individual segments, and top-ranked segments are selected to generate the final video summary. We evaluate the proposed method on The SumMe Video Summarization (SumMe) dataset and show that our method is able to achieve summarization accuracy that is comparable to that of a current state-of-the-art deep learning method, while posting significantly faster run-times. Our method on average is able to generate a video summary in time that is shorter than the duration of the video.

show abstract

To Click or Not To Click

Cited by 70 publications

References 36 publications

Beautiful and Damned. Combined Effect of Content Quality and Social Ties on User Engagement

Beautiful and Damned. Combined Effect of Content Quality and Social Ties on User Engagement

Combining Adversarial and Reinforcement Learning for Video Thumbnail Selection

Real-time Video Summarization on Commodity Hardware

Contact Info

Product

Resources

About