Machine Dreams of A Cosmopolitan Past: An Image Generator Based on Safavid-Era Iranian Miniatures

by: Hamidreza Nassiri & Rowena Chodkowski

Presented at: MUTEK Forum & Festival, Montréal, Canada, August 2024.

We trained a text-to-image Generative AI model natively with a large selection of Iranian miniatures from the Safavid era (16th to 18th century). We hand-scanned or digitally extracted these images from different books and cropped them to make them suitable for training. Then, we processed and tagged them manually. The model was then trained with the Kohya SS native training script on a cloud GPU instance.

We then converted the model and set it up to work with the Hugging Face Diffusers library.

Urban Video Archive

by: Hamidreza Nassiri, Jacob Geuder, Mídia Independente Coletiva

Presented at: The Radical Film Network Conference, Madrid, June 2024.

The Urban Video Archives Rio project analyzes how social transformations are chronicled by digital video making and how those videos can be preserved and structured for scholars, activists, and the public. These questions are of urgency in Brazil, ten years after digital video activism blossomed. Today many videos by citizen journalists and video activist collectives are in danger of being lost. The Urban Video Archive preserves digital videos and ensure their long-term accessibility in an online repository. The prototype of the digital video archive is developed in close collaboration with Mídia Independente Coletiva from Rio de Janeiro. The collection consists of 119 videos that document or bear witness to Black, Indigenous, and other marginalized communities’ struggles for human rights, and amplifies the voices of marginalized urban citizens. The project is currently supported by Fordham University’s Center for Community-Engaged Learning. We are also in the process of securing Princeton University’s library server as a safe and open access repository for the videos. The archive will be expanded in collaboration with other media activists from around the world in the hope to establish a global database of videos documenting social movements and matters related to social justice.

Global Cinemas Resource

Co-Created With Students at Fordham University, NYC

My students in Global Cinema (Fordham University) and I co-created this resource for global cinemas in Spring 2024. 

Upon clicking on any country’s name, you will see a brief overview of its national cinema, five significant films representing it with brief descriptions of them and their importance, and finally five readings on the national cinema and/or the films. 

I hope this resource will help teachers and scholars of global cinemas.

As you see, each country’s entry is written by two students. As such, the quality of the content varies a bit, but my hope is to work with some students later to edit the entries. 

Justice in Context: Making Sense of ‘Context’ in Discourses Surrounding Viral Videos of Social Injustice

Presented at the Bearing Witness Conference, Massachusetts Institute of Technology (MIT), 2022. (Co-authored with Rowena Chodkowski)

“Context” has become a buzzword on social and legacy media for fact-checking or discussing the meaning of shared information. When a video recording of an incident becomes viral, journalists and members of the public use “context” to frame narratives surrounding the event. Some discourse participants say, “this is taken out of context,” or claim “we should look at this in the context of” a larger social issue, while others share new material for “additional context.”

We Scraped a dataset of 117, 026 comments from specific threads on Reddit discussing the issue of “context” in understanding and interpreting the videos documenting the Kyle Rittenhouse shooting event.

We used quantitative and qualitative analysis to understand how people use “context” in different ways to shape their understanding.

Mapping NYC WiFi Hotspots

by: Hamidreza Nassiri

Data Sheet: NYC OpenData

I used the data collected by NYC OpenData for 3,319 Free and Limited Free WiFi Hotspots throughout NYC to create this interactive map. The user can use the map and its markers to locate hotspots and/or filter the hotspots based on the borough, provider, and type (free or limited free).

Access Here

Civic Media Workshops

by: Hamidreza Nassiri, Ralph Vacca, Jessica Baldwin-Philippi, Gregory Donovan

Over the course of a year, I led a team of faculty members in Fordham University’s Media & Communication Studies department to organize Civic Media Workshops, aiming to incorporate civic media with a focus on racial justice into the curriculum and student learning experiences. Each civic media workshop was facilitated by an artist/activist and focused on a specific project and media-making approach. This project has become possible thanks to the generous grant from Fordham University’s Office of Chief Diversity Officer.

Workshops’ recordings and accompanying resources are accessible to the public via https://civicmedia.ace.fordham.edu.