# INDRA Project Report Template --- ## Preliminary MK/MW Notes * MW: * Indra notes from her LLUF application: * "I am a sophomore concentrating in Computer Science. I am very interested in artificial intelligence, its potential uses, and how it intersects with artistic and creative processes as I also love visual arts. I am also interested in exploring natural language processing since I am passionate about languages. I am proficient in Mandarin Chinese, Spanish, and French. At the Learning Lab I would like to get more experience with coding. My experience in programming includes Python, SQL, R, HTML, CSS, and Flask. I have experience with image processing, face and object recognition, and processing and analyzing large quantities of financial data. I have experience with libraries such as OpenCV, NumPy, Pandas, Matplotlib, among others. I would especially like to do projects using TensorFlow, or collaborate on projects to learn more about React and Next JS." * Indra project: * based on her interests, I feel like Indra could work primiarly on our internal workflows and camera vision. She has a lot of experience in this area, and I think she could then help bridge those workflows to our larger "machines" for turning studio work into text/code/products/etc. --- ## Project Description ### Project/Tool Overview **Project Name:** **Associated Harvard Course:** **Summary:** _Describe the project/tool, its purpose, and how it fits into the course._ **Technology Stack:** *What tools are you using to craft your project? OpenAI APIs? Slack APIs? Python Notebooks, next.js apps, etc?* ### Learning Objectives and Benefits **Objectives:** _List the specific learning objectives this tool aims to address._ **Benefits:** _How will students or professors benefit from using this tool? Describe the expected impact on teaching and learning. --- ## GAICA mk check-in **Date:** 20240228 * Marlon wants to brainstorm an initial project: * Indra pitched creating a CV tool that identifies faces from video, specifically the faces of people who work here * then stills are captured, saved somewhere, and printed * marlon likes this for an "initial learning project"/proof-of-concept * he wants her to define her specs, then create the code * Indra brainstorming * marlon wants her to ground her specs in "why are we doing this?" * why are we identifying faces and movement? * for example, being able to identify faces in footage and use it as a way to define and categorize video clips * how abouts cards/art supplies? * marlon tells indra he wants more tools that identify cards/paper from an overhead cam * colors, text, movement, order/composition * for example: * if indra can create a tool that identifies the color/content of a card coming on the overhead cam, and then connect that to various capabilities (changing lights, triggering sound, etc.) that would be both useful and "magical" * very useful for workshops * **as a result, this code must work with "streaming" video, so we can do this (eventually) live.** * marlon shows OBS, to demonstrate it as the tool we use for live streaming * recognizing cards * this is decided to be the first project, starting by: 1. creating a tool that processes extant videos and recognizes that: * there is a card * a still is taken * the color of the cards * extra add ons could be: * text * shape * motion * composition * rotation/orientation 2. next step would be to do this w/ live streaming video 3. after this, indra can build out capabilities * still frame of cards taken, saved to database * the categorization of this database should remain flexible and focsed on getting as much information as possible in an "un-opiniated way." * functions/role of this data can be decided later * certain cards triggering certain functions * ligth changes, sounds, etc. * techstack * the code will be hosted in github * we'll get her general information and the info for the public repository * slack channel * currently for posting information * but may scale up to creating slack bots * the slack bots could be the way in which the cards act as triggers to extra functions and vice versa. * APIs/APs * Indra asks about use * marlon says we should default to large, well-maintained libraries and tools * free is also a bonus * however, APIs are an options as well * just check in with marlon first * database * airtables * action items: * needs for [doc](/uhnr51hFTBy0xtRy0Md2Mg): * videos of cards, stills, etc. * specs of cards --- ## Bi-Weekly Update **Date:** [Submission Date] ### Interesting Developments _Describe any interesting occurrences, breakthroughs, or anecdotes from the past two weeks._ ### Resource Requests _Identify any additional resources or support needed._ ### Best Practices & Learnings _Share best practices discovered or lessons learned during the project implementation._ ## Collaboration and Feedback ### Staff Collaboration _How have you collaborated with course heads or teaching staff in the past two weeks?_ ### Feedback Implementation _How has feedback been utilized to iterate on the project/tool?_ ## IV. Future Directions ### Next Steps _What are the immediate next steps for the project?_ ### Long-Term Vision _Where do you see this project going in the future?_ ## VI. Additional Notes ### Miscellaneous Information _Any other information or personal notes that may be relevant to the project._ ---