Salienteye: Maximizing Engagement While Maintaining Artistic Style on Instagram Using Deep Neural Networks

  title={Salienteye: Maximizing Engagement While Maintaining Artistic Style on Instagram Using Deep Neural Networks},
  author={Lili Wang and Ruibo Liu and Soroush Vosoughi},
  journal={Proceedings of the 2020 International Conference on Multimedia Retrieval},
Instagram has become a great venue for amateur and professional photographers alike to showcase their work. It has, in other words, democratized photography. Generally, photographers take thousands of photos in a session, from which they pick a few to showcase their work on Instagram. Photographers trying to build a reputation on Instagram have to strike a balance between maximizing their followers' engagement with their photos, while also maintaining their artistic style. We used transfer… 

Tables from this paper



How to Become Instagram Famous: Post Popularity Prediction with Dual-Attention

A novel dual-attention model is devised to incorporate image, caption, and user environment and shows that the model outperforms the baselines and a statistical analysis identifies what kind of pictures or captions can help the user achieve a relatively high "likes" number.

Faces engage us: photos with faces attract more likes and comments on Instagram

The first results on how photos with human faces relate to engagement on large scale image sharing communities are presented, finding that the number of faces, their age and gender do not have an effect.

What makes an image popular?

The importance of image cues, such as color, gradients, deep learning features and the set of objects present, as well as the importance of various social cues such as number of friends or number of photos uploaded that lead to high or low popularity of images are shown.

Multimodal Popularity Prediction of Brand-related Social Media Posts

Experiments on a collection of fast food brand-related user posts crawled from Instagram show that visual and textual features are complementary in predicting the popularity of a post, and predicting popularity using the proposed engagement parameters is more accurate than predicting popularity directly from visual and text features.

Cats and Captions vs. Creators and the Clock: Comparing Multimodal Content to Context in Predicting Relative Popularity

The experiments show that when considered in isolation, simple unigram text features and deep neural network visual features yield the highest accuracy individually, and that the combination of the two modalities generally leads to the best accuracies overall.

User-guided Hierarchical Attention Network for Multi-modal Social Image Popularity Prediction

A model named User-guided Hierarchical Attention Network with two novel user-guided attention mechanisms to hierarchically attend both visual and textual modalities is proposed, capable of not only learning effective representation for each modality, but also fusing them to obtain an integrated multi-modal representation under the guidance of user embedding.

Learning and Transferring Mid-level Image Representations Using Convolutional Neural Networks

This work designs a method to reuse layers trained on the ImageNet dataset to compute mid-level image representation for images in the PASCAL VOC dataset, and shows that despite differences in image statistics and tasks in the two datasets, the transferred representation leads to significantly improved results for object and action classification.

Understanding image virality

  • Arturo DezaDevi Parikh
  • Computer Science
    2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2015
This work trains classifiers with state-of-the-art image features to predict virality of individual images, relative virality in pairs of images, and the dominant topic of a viral image, and compares machine performance to human performance on these tasks.

"Nobody comes here anymore, it's too crowded"; Predicting Image Popularity on Flickr

This paper predicts the number of comments and views an image has based on a number of new features for this propose, and is able to overcome the problems associated with popularity prediction in a cold start.

Rethinking the Inception Architecture for Computer Vision

This work is exploring ways to scale up networks in ways that aim at utilizing the added computation as efficiently as possible by suitably factorized convolutions and aggressive regularization.