NM3213 Final Project: Filtering Instagram

Please refer to prototype link here: https://invis.io/K399G78SF

Please refer to prototype presentation in Google Slides here: https://drive.google.com/open?id=0B1RR1M9PfZi7NmZzNDFNRkZ1Qmc

1. Introduction

Named as ‘Filtering Instagram’, this project is an interactive webpage with a moving crossword puzzle in the home page that highlights the crowd wisdom adopted across different countries/cultures in terms of aesthetic styles in photographic editing. It also encourages users to create a short, non-linear still image film made up of images with the same filter. Users will be able to adopt a simple ‘drag-and-drop’ function at the Create tab to create their own narrations out of those visuals available.

The concept from this project stems from the increasing reliance on the social media platform, Instagram, for social currency. In today’s social networks, it is appalling to believe that Instagram, an online mobile photo-sharing, video-sharing, and social networking service can hold such gravitas in our society – with 400 million active users as of April 2016, (Chaffey, 2016). This project seeks to rethink how Instagram is, or can be used as a form of social performance to display identities and statuses by observing the various kinds of filters used in different photo categories and/or users across different geographical regions. Future developments involve placing the photos in a broader context of photography, investigating how photos uploaded to Instagram differ in terms of its aesthetic styles in different geographical regions. Any plausible patterns and/or phenomenon should be viewed in context with the history of photography, and the increasing rise of technological developments that have led to easy availability of image editors and/or manipulators.

2. Literature Review

2.1 About Instagram

Instagram, a mobile photo (and video) sharing, social networking service that enables its users to take pictures and/or videos before sharing them either publicly or privately on the application, has quickly emerged as a popular medium among young adults in recent years. Its ability to connect through a variety of other social networking platforms, such as Facebook, Twitter, Tumblr, and Flickr, provides users an instantaneous way to capture and share their life moments with friends. Users can choose to manipulate their photos with in-app editing tools which include numerous different colour filters that are superimposed on the photos.

From a study conducted, it was revealed that the Instagram photos can be roughly classified into eight different types of categories based on their content. The categories are namely self-portraits (i.e. selfies), friends, activities, captioned photos (i.e. pictures with embedded text), food, gadgets, fashion, and pets (Hu, Manikonda & Kambhampati, 2014). Intended for sharing photos on-the-go with the increasing smartphone use, features expanded into allowing users apply different manipulation tools before posting; add captions, hashtags using the ‘#’ symbol to describe the media, and tag or mention other users by using the ‘@’ symbol. With 40 billion photos shared thus far, the average number of Instagram posts per day per user profile is estimated to be 2.69 posts per day (Smith, 2016). These statistics highlighted the significance of this project, as well as the presence of a readily-available database to be tapped on. Having deep understanding of Instagram allows a deep understanding about social, cultural, and environmental insights about people’s activities – understood through the lens of their photos. As compared to other text-based communication platform, the level of intuitiveness is higher since different interpretations of similar issues by individuals can be explored and discovered.

2.2 Instagram as Database

Being a visual-locative social medium, Instagram serves as a huge database with users producing data daily as they navigate their daily lives, documenting every key moment by capturing them and posting online. Defined by Ramsay (2004), database is a system that allows for the efficient storage and retrieval of information. As such, Instagram lends itself as an unparalleled data source that can be explored to discover socio-spatial patterns and divisions across different geographical regions. It allows individuals at-large to collaborate and make use of the information in ways that disregard institutional boundaries. Also, mentioned in the article, where complications arise with the issue of organization and efficient retrieval of data due to the need to facilitate interaction with multiple end user and allow for dynamic insertion and deletion of information, Instagram’s classification was facilitated with the use of hashtags.

Photos have limited power in how they would be interpreted – which explains the need for contextual information relayed through captions and/or from user profiles posting them. Users attribute specific tags to their photos in their captions which allow their photos to be categorized under the community-generated classifications. Aimed at cataloguing photos under various themes for more lasting impact, hashtags help to describe the image with a limited number of words, enable users to easily locate other similar photos taken, and thus makes it easier to retrieve, use, or manage the information resource. This is in line with the purpose of metadata, as asserted by the National Information Standards Organization (2004), where it is commonly used for any formal scheme of resource description – in the case of Instagram, it falls under its caption. Allowing for easy search of innumerable photo categories and/or posts by relevant criteria, this feature also enables the recognition of (underlying) photo content by providing the context. For instance, a hashtag can also serve as a symbol of a community membership that breaks down the virtual barriers between users (Yang, Sun, Zhang, & Mei, 2012).

2.3 Posting as a form of Social Performance

In Manovich’s paper titled “Subjects and Styles in Instagram Photography (Part 1)” (2016), Martin Hart described social network photography as “visual publicization of ordinary life in a ubiquitous photograph” and that they are just “strategic imagery” aimed to persuade or promote a certain social performance. This idea also translates into identity formation when the pictures (i.e. data) available, shared by users themselves, can be used to identify distinct groups of users, shed light on plausible segmentation and social statuses. However, point to note that the information portrayed despite being up-to-date, may not be the most comprehensive, or simply reflecting the activities since Instagram users can selectively represent their worldviews by displaying pictures they deem fit and suited for circulation. Additionally, individuals across different geographical regions may understand and use the Instagram medium differently. Often so, the subjects taken and photographic styles in the photos, are strongly influenced by the social, cultural, and aesthetic values of a given location and/or demographics.

Likewise, proclaimed by Manovich, the analysis of Instagram content as social performance in term of its visual aesthetics is crucial as different meanings may be conveyed depending on how they are being photographed. With different interpretations, they may either look like random things in a background, or subjects in a sophisticated, purposely-styled composition. For meaningful findings, analysis of the visual aesthetics alone cannot rely plainly on the subjects, but along with the user’s intentions and other contextual information (as mentioned before). For example, a photo with an unbalanced composition may perhaps be unintentional, signalling a careless photographer, or represent the work of an abstract Instagram author making a very conscious decision on the aesthetics portrayed.

Individuals are exposed to thousands of images every day through advertising, and then Instagram on top of that – how is one not feeling discomfort? This is because individuals are not seeing every one of those images, but rather image types. Users are enjoying the variation between types instead of seeing a full image each time (Manovich, n.d. as cited in Thomsen, 2016). Bearing that in mind, individuals’ daily encounter with large amounts of data creates a new “data subjectivity” – each with its own unique blend of imagination. The concept of ‘soft cinema’ looks at novel ways of visualizing this subjectivity. In short, ‘soft cinema’ emerges from contemporary patterns of production and/or consumption based on the digital age and seeks to explore the convergence of database and narrative. Since the formation of self-identities have been continually scattered across multiple databases (i.e. representing the cross-platform social media mediums adopted), ‘soft cinema’ symbolized a more revealing metaphor for understanding contemporary identity rather than a traditional, linear, chronological form (Anderson, 2006).

2.4 Aesthetics Issues of Instagram: Filters

After exploring the idea of social performance and identity through Instagram, there is a need to understand the aesthetics of the platform itself. Photographs posted are presented as a continuous stream of images from numerous users based on one’s following (i.e. newsfeed). Under an individual’s account, they are visualized in the form of a gallery. When a guest first enters the gallery, he/she can see six side-by-side images in square grid – getting an overview of the user’s photographic style and an understanding of the type of content he/she is specialized in posting (i.e. under which major themes). To be able to signal their “identities” in different genres, users adopt different Instagram grids to suit their theme. And it is with filters, people can manipulate how their photo grid will look like, to ensure a cohesive aesthetic look. Filters being one of the pre-attentive aesthetic attributes that a user will notice first, it is with this certain recognizable way of expression that forms part of the contemporary visual culture to strategically present imagery in a form of social performance. This explains the need for “cohesiveness” and consistency in Instagram editing style to allow the feed flow naturally.


Refer to link: https://www.flickr.com/photos/culturevis/27064111290/sizes/l.

According to a study from Yahoo Labs, researchers analysed 7.6 million photos and came to the discovery that “filtered photos are 21% more likely to be viewed and 45% more likely to be commented on” (Mchugh, 2015). The same study also revealed that 76% of photos has undergone image processing and out of the remaining 24%, it includes pictures which has been edited out of Instagram and thus did not adopt the native filters. Not only that, results show that individuals are more attracted to images with higher contrast and corrected exposure, and find a warmer temperature more appealing than a cooler one. By adding hues, grain, contrast and so on, including filters and/or other manipulation tools, it can evoke a different ‘feel’, thus changing the message communicated by the picture. For instance, by adding filters like ‘1979, a different time period may be suggested from the colour hues overlaid on the picture. This results in a multi-temporal image that can suggest different temporal references where each can tell a different narrative on its own when placed in different context.

The motivation behind adding photo filters are clear – to alter the feel and meanings of the photos that will in turn represent a false reality. However, such distortion appears to be acceptable by users with the increasing popularity of users making use of filters, coupled with the proliferation of external filter-adding applications like VSCO, RNI Films and so on. Recently, there was even an app, named Prisma, that popularized the art filter craze by allowing photos to turn into artworks by superimposing styles of renowned artists, famous ornaments, and patterns.

Such is what Fernandez (2014) calls the “new old” phenomenon – where socially recognized styles are applied to the photos create a vintage look to images by playing with the colour, or fading certain areas, and are manipulated to fit modern aesthetics. She also quoted Fredric Jameson’s (1991) “nostalgia for the present”, arguing that postmodernity is not concerned with historic elements but rather imitates a version of the past to control the present consumer culture. Photo filters create a hyperreal representation of the world since the photos may not have any basis in reality, but nonetheless posted online anyway and made to believe they represent reality itself. This reflects the symbolism of postmodernism since it becomes “more real than the real” (Sturken, 2001) and having such impression of history adds to the significance, which perhaps explains the reason pictures with a more saturated note (e.g. sepia) are not receptive to ‘likes’ (Tossell, 2012).

Following that, this project aims to investigate if the Instagram photo filtering or visual editing culture is disparate or unique across different subject themes. Additionally, how can the findings uncover trends between different cultures considering social influencers are advertising and sharing the types of editing styles they use for their pictures. In answer to that, there had already been a branding research done in 2016 to uncover ‘the most popular Instagram filters from around the world’ that highlights the different cultural analytics in data visualization aesthetics. The findings also revealed the most-used filters for various types of subject themes. It will be illuminating to see their findings, refer to the link here: https://designschool.canva.com/blog/popular-instagram-filters/.

2.5 Related Work

In this section, related work done by different artists and corporations will be looked at to draw plausible inspiration for the final idea conceptualization.

Firstly, Lev Manovich, Nadav Hochman and Jay Chow’s project titled Phototrails was chosen since it aimed to explore social media’s visual data and gather plausible insights from the massive amounts of user-generated content. It is an urban representation of cities by people who capture and share photos on social networks that highlighted the patterns based on geographical difference and activity levels – focusing on spatial and temporal dimensions. Ultimately, high resolution visualizations were created with custom software using 2.3 million Instagram photos from 13 global cities showing the temporal changes in the photos, locations, and visual characteristics which can uncover social, cultural, and political insights about people’s activity around the world. The project being part of an emerging field of cultural analytics, used computational methods for the analysis for the massive cultural datasets and flows. I was able to draw inspiration from this project – especially how they analyse each photo closely, down to even their metadata, so as to read the “stories” made up by the users’ sequences of photos.


Secondly, Metagramme, a side project by T2D agency, was created to explore the patterns that can be created or emerged by combining visual data over time. It is an interactive website that extracts 36 or 64 of recent Instagram photos and combines them into a single image which will result in a colourful digital amalgam that is part photography and part abstract art as the averaging of photos creates a composite image that yields a blurry effect. Also, adopting the idea of data categorization with hashtags, the website can also pull information from specific hashtags to show what a blend of 64 photos of a specific subject looks like. I like the idea of how Metagramme is essentially using data to create a visualization, but not for practical uses, rather, it is to create something beautiful. I also draw inspiration from their concept of representing emotions like “love” or “sadness”, which can be challenging – but the creation of the abstract art helps explore how different people conceptualize a specific issue, or represent them over time. Refer to their webpage here: https://metagram.me/,

Thirdly, another collaborative work by Manovich, he aimed to discuss with Tifentale the construction of popular photographic self-representation in digital visual culture. It analysed 3,200 selfies shared via Instagram from five global cities namely Bangkok, Berlin, Moscow, New York and Sao Paulo. Albeit the focus point of this project ties down to a specific type of imagery produced, I felt that it ties in with the cultural differences that can be attributed based on different representations for social performances across cultures. This project made use of image analysis as well, to place the selfie into a broader context, and hence argue that this is a new sub-genre of photography that differs or possibly evolves from traditional self-portraiture. The result also showed how the methods of self-portraiture change over time with technology and the rise of smart phones – similar to what I would want to bring across about the point of photo editing.

selfie 1.png

selfie 2.png

Lastly, mentioned under the segment for ‘Posting as a form of social performance’, I also seek to explore the idea of social media database since Instagram can be a platform for massive amounts of user generated content. Such media elements from the database can used by various users of different perceptions to construct a potentially unlimited number of narrative films. This is taken reference from Soft Cinema where it makes use of a mode of dynamic, real-time image assembly, executes a series of choices, guided by carefully designed parameters and rules (algorithms), in order to deliver a narrative experience that varies each time it is played. The result is a kind of “ambient narrative” in which narrative meaning and aesthetic coherence must be discerned or constructed by the viewer.

3. Proposed Idea

3.1 Process of Devising

Bearing the research done and influences in mind, my initial research questions were as follows:

A. How does Instagram shape photographic art, especially with the proliferation of smartphone photography (i.e. iphonegraphy)?

B. How do the visual characteristics of pictures differ across different users posting about different genres/issues?

C. What is the potential of social impact with the use of Instagram?

However, feedback received for this initial pitch was that there was no specific focus and my research questions were not hypothesis I was trying to prove, but assumptions, since there was no literature to back them up.

Hence, drawing from my literature review, I pivoted and devised a new set of research questions focusing on the Instagram filters used. I aim to find out:

A. How are the visual characteristics of pictures different across different user groups of different themes?

B. What is the potential of Instagram database visualization and storytelling?

These research questions serve to explain some of the underlying theoretical concepts highlighted from the literature review. For instance, aesthetics issues of data visualization can be uncovered through the conscious choice of users to use specific filters over the others to form the pre-attentive attributes of the imagery. It was also aimed to uncover the potential of the database, in the form of a social network, to see how researchers can work with the massive amounts of user-generated content to develop valuable insights with regards to digital humanities, rather than just for corporate use. Lastly, since the overarching theme of the paper talks about Instagram posting as a form of social performance and the use of photo filters are a form of representation of users’ own interpretation of reality, the last research questions seeks to uncover the potential of storytelling by converging the Instagram database together with non-linear forms of narrative with the various captions and hashtags that users have posted.

3.2 User Experience


Refer to the user experience task flow diagram as shown above.

Essentially, it will be an interactive webpage with a crossword puzzle made up of photos of different genres and/or subject themes. This is to symbolize the documentation of lives with photo sharing as a fundamental public act, with users organizing themselves in different communities around photographic styles/subjects. Users can zoom in to photos of the same genre being grouped together to see the various filters used and other meta data like the other editing configuration. Being able to zoom in to view the aesthetic effects allow users to comprehend the cognitive and emotive styles conveyed and how they fit into the broader trajectories of visual culture.

Additionally, there is another key feature of the web page where users can decide to create their own narratives. In that page, the gallery feed available for users will be all the same filters. Users can use the simple drag-and-drop function to form short films making up of images with the same filters. After which, they can choose to edit the current captions and/or add their own to ensure the narrative flow. This is to symbolize the various conceptions individuals have of the same worldview and it will be interesting to note the infinite amount of arrangements the photos can be re-arranged to create narratives to convey different meanings. With existing data from the database and digital archives (which consists of not just the primary data, but also other experiences users want to convey via the pictures), there is plenty room for development in terms of data visualization to narrate their stories. After creating their own short still-image film, users can also share to various social media networks.

4. Limitations and Future Developments

Instagram, like all other major social networks, possesses an API that allows anyone to download some of its pictures and selected associated data. An individual can choose to download images by location, by tags, or by specific usernames. For each image, metadata that can be retrieved include its tags, description (i.e. caption), date/time stamp, geo coordinates, username, list of users who had liked the image, and comments. There is a challenge in visualizing such high-dimensional data considering the various possible readability and interpretations. Categorising them is also one of the difficulties faced in this project which thus highlights the importance of setting parameters. Not only that limitations to execute this project also involves the technical specifications of it. Considering the profusion of visual data being captured by Instagram, such big data requires specialized technical skills to collect and analyse. To create the interactive webpage alone with the moving visuals requires programming skills which sadly, was not the focus of this module. Hence, for future developments, there is a need to consider this as well.

Future developments can also perhaps be considering data from external photo editing applications like VSCO, Snapseed, Photoshop and so on. Additionally, instead of developing own narratives, users can generate mood boards with fitting filters according to how they are feeling so they (or others) can refer to them for inspiration. This ties in with Metagramme’s notion of conceptualizing intangible tags like emotions. Further development can also be made by including text analysis to include the users’ initial emotions and rationale for posting the posts they did and creators can incorporate that into their narrative writing on the web.


Anderson, S. (2006). Soft Cinema: Navigating the Database (Review). The Moving Image. Volume 6, Number 1, Spring 2006, pp 136-139. Minnesota: MIT Press

Chaffey, D. (2016). Global social media research summary 2016. Retrieved from http://www.smartinsights.com/social-media-marketing/social-media-strategy/new-global-social-media-research/.

Fernandez, N. (2014). Filtered Realities: Instagram, Photo Filters, and Postmodern Images. Retrieved from http://www.itaintnecessarilyso.org/articles/2014/08/22/filtered-realities-instagram-photo-filters-and-postmodern-images.

Hu, Y., Manikonda, L., & Kambhampati, S. (2014). What We Instagram: A First Analysis of Instagram Photo Content and User Types. Retrieved from https://www.aaai.org/ocs/index.php/ICWSM/ICWSM14/paper/viewFile/8118/8087

Jameson, Fredric. 1991. Postmodernism, or, the Cultural Logic of Late Capitalism. Durham: Duke University Press.

Manovich, L. (2016). Subjects and Styles in Instagram Photography (Part 1). Retrieved from http://manovich.net/content/04-projects/090-subjects-and-styles-in-instagram-photography-part-1/lm_instagram_article_part_1_final.pdf.

McHugh, M. (2015). Turns Out There Are A Lot of Academics Studying Photo Filters. Retrieved from https://www.wired.com/2015/05/instagram-filters/.

National Information Standards Organization, “What is Metadata?” (Bethesda, MD: NISO Press, 2004)

Smith, C. (2016). 180 Amazing Instagram Statistics. Retrieved from http://expandedramblings.com/index.php/important-instagram-stats/7/.

Stephen Ramsay, “Databases,” in Companion to Digital Humanities, edited by Susan Schreibman, Ray Siemens, and John Unsworth (Oxford: Blackwell Publishing Professional, 2004)

Sturken, Marita and Lisa Cartwright. 2001. Practices of looking: An introduction to visual culture. Oxford: Oxford University Press.

Thomsen, M. (2016). What’s Instagram Doing to Us? Retrieved from http://www.complex.com/life/2016/03/lev-manovich-interview-instagram-project.

Tossell, I. (2012). The Curious Appeal of the Instagram Filter. Retrieved from http://www.theglobeandmail.com/technology/digital-culture/the-curious-appeal-of-the-instagram-filter/article4179844/.

Yang, L., Sun, T., Zhang, M., M, Q. (2012). We Know What @You #Tag: Does the Dual Role Affect Hashtag Adoption? WWW 2012 – Session: Behavioural Analysis and Content Categorization in Social Media. Vol. April 16-20, pp 261-270.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s