Differences

This shows you the differences between two versions of the page.

--- newsletters:2026-01 [2026/02/02 23:38] – [OCR experiments] osnr
+++ newsletters:2026-01 [2026/02/10 01:19] (current) – Add projector focus adjustment photo cwervo
@@ Line 1: / Line 1: @@
 ====== January 2026 newsletter ======
-Our next [[https://TBD.com|Folk open house is in the evening on ????, February ??th, in East Williamsburg, Brooklyn.]]
+Our next [[https://luma.com/ptzbza54|Folk open house is in the evening on Thursday, February 26th, in East Williamsburg, Brooklyn.]]
 ===== What we've been up to =====
-  * Daniel esc-pos printing fix
+  * [[https://danielpipk.in|Daniel Pipkin]] [[https://github.com/FolkComputer/folk/commit/8551cff340862ded01e57ec6e67d790294f5e823|fixed our ESC/POS receipt printer support for folk2]]
-  * Esben Sørig camera/setup fixes
+  * [[https://github.com/sorig|Esben Sørig]] contributed some camera/setup fixes:
+    * [[https://github.com/FolkComputer/folk/pull/241|Extend camera load timeout to 2 seconds and add retry mechanism]], which seems to fix the Logitech C270 camera
+      * Also don't die on non-fatal JPEG decompress errors
+    * [[https://github.com/FolkComputer/folk/pull/242|Make aspect ratio of camera preview on /calibrate adapt to actual camera aspect ratio]]
+  * Andrés and Brian worked on the shape library motivated by ideas for drawing shapes to highlight objects on the table
+    * {{.:pasted:20260210-001649.png?400px}}
+    * Preview of the simplified shape code working in folk2 to draw a square on a quad
 ==== OCR experiments ====
+Omar: I've been wanting to do non-AprilTag inputs for a while, especially handwriting input ([[newsletters/2025-01#handwriting-recognition-proposal|see proposal from last year]], for example).
+I started looking at OCR/handwriting recognition more seriously a couple of months ago, and [[newsletters/2025-11?#trocr|I concluded that TrOCR was probably the right building block (AI model) for this.]] It runs pretty fast (0.1s to 1s) on most systems on GPU and CPU, I was able to run it locally, it's trained for handwriting and not print. I did some prototyping in Python for that.
+Now that we have folk2, it's a lot easier to run synchronous tasks like handwriting recognition without blocking the rest of Folk, so I want to finally get stuff like this deeply integrated (and have the integration be pretty simple and idiomatic) and prototype how to interact with it.
+So at the end of January, I finally started hacking on a TrOCR integration that runs inside Folk, so we can actually write on the table and get live text output.
 {{newsletters:img_1463.jpeg?400px}}
 ==== Stuff from China ====
+Omar:
 {{newsletters:img_1439.jpeg?400px}}
 {{newsletters:img_1432.jpeg?400px}}
+{{.:pasted:20260210-011824.png?400px}}
 TODO: table one
 ==== Outreach ====
@@ Line 29: / Line 48: @@
 ===== What we'll be up to in January =====
-  * Our next [[https://TBD.com|open house is on ????, February ??th]]
+  * Our next [[https://luma.com/ptzbza54|Folk open house is in the evening on Thursday, February 26th, in East Williamsburg, Brooklyn.]]
+  * Andrés: Finishing shape library
+  * Andrés: Working on restoring and making demos with the dot detector capability
 ===== Links we've enjoyed =====
@@ Line 44: / Line 65: @@
   * https://xkcd.com/color/rgb/
   * https://www.youtube.com/watch?v=9ATpyhBWoPk
 ==== Andrés ====
+  * [[https://bitsavers.org/pdf/xerox/alto/|A large collection of PDFs from Xerox PARC]]
+  * [[https://iitcoa3rdyr.wordpress.com/wp-content/uploads/2013/03/allen_diagrams-matter.pdf|Diagrams Matter]] by Stan Allen
+  * [[https://github.com/vikhyat/moondream|Moondream]] a tiny vision + language model for running on things like Raspberry Pis