Skip to content Skip to navigation

Lightning Talks with CIDR

October 27, 2020 - 12:00pm

Please join us for a virtual CIDR-CESTA lightning talk gathering. This is an informal event geared towards bringing together researchers working on DH projects to allow them to share ideas and challenges and to gather feedback and insights from peers. Short presentations on work-in-progress will be followed by open discussion. Join the virtual gathering via Zoom.

DH Lightning Talks on October 27

Social Networks and Geographical Mobility in the Han Empire
Yunxin Li (Graduate Student, History)

I will introduce how I have used social network analysis to study the politics of the Han empire, discuss my current work on the geographical mobility of Han officials, and finally talk about how I plan to combine network analysis with spatial data.

Mapping Geographical Knowledge in the Middle East

Adrien Zakar & Merve Tekgurler (History)

Hundreds of geographical books and illustrations have been produced in Arabic and Ottoman-Turkish since the late 18th-century. This body of spatial knowledge comprises both geography, information concerning life on earth usually portrayed as points, lines, and polygons, and cosmography, theories about the earth’s position in the universe represented as circles. The proliferation of geographical language and tools two centuries ago reflected the rise of the territorialized nation-state, intensifying militarization, and sweep capitalization of Middle Eastern economies. Atlases, compendiums, and treatises are an invaluable source for studying these processes as they unfolded in the MEASA region. Despite the value of such a corpus, its sheer scale has long deterred researchers from exploiting its evolution across time. This project aims to offer an open archive that gathers together the extent of modern geographical works into a user-friendly map comprised of three main layers which will visualize where each work was produced, what locations they depicted, and where these works can be found today. This map builds on a dataset of geographical publications in Arabic and Turkish, with occasional insertions of relevant materials in other languages including Urdu, Persian, French, English, and German. These include works explicitly titled, “geography,” “cosmography” and “atlas” as well as a plethora of vernacular genres in spatial analysis such as "Salname" [Almanac] “Khiṭaṭ” [Plans] “Cihannuma” [Book of the Earth] and “Kitab al-ʿAjaʾib” [Book of Wonders] among others. The dataset builds on existing printed repositories of references, such as Ekemeleddin Ihsanoğlu’s two volume “Osmanlı  Coğrafya Literatürü Tarihi” (History of Ottoman Geographical Literature).

Comparative K-Pop Choreography Analysis through Deep-Learning Pose Estimation across a Large Video Corpus

Peter Broadwell (Digital Scholarship Developer, CIDR)

K-pop dance performances, particularly those of corporate and government-sponsored “idol” groups, are a key component of K-pop’s core mission of projecting soft power internationally. Consequently, thousands of K-pop music videos and official “dance practice” recordings are readily available via online video platforms, where they often garner millions of views. While there are well-known choreographers who work with K-pop production companies, there is little documentation of the choreography beyond the dance videos themselves. The recent advent of deep learning-based pose detection methods that can reliably detect human body/limb positions from video frames makes possible computational studies that use these videos as primary sources for “distant viewing” (as well as close viewing) analyses. This talk outlines work done to date to detect and quantify synchronized poses, concerted motion (“tightness”), and recurrent poses and motions in specific videos, contributing to the development of a typography of K-pop dance idioms that can facilitate a data-driven study of the constitutive interdependence of K-pop and other cultural genres.

For more information or if you're interested in presenting in a future session please contact Marie Saldaña (saldana@stanford.edu).

 

Event Sponsor: 
Center for Spatial and Textual Analysis (CESTA) and CIDR
Contact Email: 
saldana@stanford.edu