
Hail, fellow adventurers: to prove I do something more than just draw and write, I'd like to send out a reminder of the Second Embodied AI Workshop at the CVPR 2021 computer vision conference. In the last ten years, artificial intelligence has made great advances in recognizing objects, understanding the basics of speech and language, and recommending things to people. But interacting with the real world presents harder problems: noisy sensors, unreliable actuators, incomplete models of our robots, building good simulators, learning over sequences of decisions, transferring what we've learned in simulation to real robots, or learning on the robots themselves.

The Embodied AI Workshop brings together many researchers and organizations interested in these problems, and also hosts nine challenges which test point, object, interactive and social navigation, as well as object manipulation, vision, language, auditory perception, mapping, and more. These challenges enable researchers to test their approaches on standardized benchmarks, so the community can more easily compare what we're doing. I'm most involved as an advisor to the Stanford / Google iGibson Interactive / Social Navigation Challenge, which forces robots to maneuver around people and clutter to solve navigation problems. You can read more about the iGibson Challenge at their website or on the Google AI Blog.

Most importantly, the Embodied AI Workshop has a call for papers, with a deadline of TODAY.
Call for Papers
We invite high-quality 2-page extended abstracts in relevant areas, such as:
- Simulation Environments
- Visual Navigation
- Rearrangement
- Embodied Question Answering
- Simulation-to-Real Transfer
- Embodied Vision & Language
Accepted papers will be presented as posters. These papers will be made publicly available in a non-archival format, allowing future submission to archival journals or conferences.
SubmissionThe submission deadline is May 14th (Anywhere on Earth). Papers should be no longer than 2 pages (excluding references) and styled in the CVPR format. Paper submissions are now open.
I assume anyone submitting to this already has their paper well underway, but this is your reminder to git'r done.
-the Centaur
















Mostly vaccine recovered, but didn't sleep well. Pretty tired, crashing out early.
Drawing every day.
-the Centaur 


















SO! Once again, I have written more than 50,000 words in a month - this time, on Dakota Frost #7, SPIRAL NEEDLE, which is close to being finished. (Yes, yes, YES, I know, Dakota Frost #4-#6 and Cinnamon Frost #1-#3 are not edited yet, editing is harder than writing, and pays less than teaching robots to learn. I'll get to them, I'll get to them, I promise). I can't figure out the new Camp Nano interface to make it cough up the usual winner banner, so you'll just get that screenshot instead.
This is my twenty-ninth victorious Nano challenge and thirty-first attempt overall. That's great stick-to-it-ness, but I was behind for much of the month, not getting my feet under me until the 10th, but I managed a big pushes two weekends a go and a huge push last weekend, leading to me briefly getting ahead of the game right around the 28th, making today an easy coast (1500 words finished me off, though I wrote through to a notch over 1,667 words just for completeness). According to my records, that 8,154 word push on the 25th was the second most I've ever written in a day, topped only by my 9,074 word mad push to finish PHANTOM SILVER, Dakota Frost #5, on July 30th, 2016.
Overall, a bit behind this month, which was pretty rough OKR (Objective / Key Result) planning at work. I love the IDEA of OKRs - say what you want to do (Objective, for example, write roughly 1/3 of a novel) and how to measure it (Key Result, for example, 50,000 words in the month of April), but this time it took us until almost the 20th. 3 weeks is way too long to spend on planning for a quarter's worth of effort.
OH, almost forgot, an excerpt:
Quick sketch of FKA Twigs. Despite my best efforts redrawing the face 2-3 times in non-repro blue, her features swam towards the bottom right of her face, and her jaw isn't angular enough. Features being good relative to each other but poor with respect to the face seem to be one of my problems. This one might be a good candidate for a trace of the picture in vellum to see the difference between the lines I drew and the lines that are actually there (insofar as lines exist in pictures, which they sorta don't).
Drawing every day.
-the Centaur
P.S. 300+ words so far, will try to push a little bit more before crashing. Only ~2700 words to go for the month.
Another quick sketch of Arnold Schwarzenegger, again roughed with non-repro blue, but this time more properly rendered with Sakura Graphic 1 and Micron 03 and 08. I was wondering if the sketch would turn out better if I properly rendered it, even quickly, using finer instruments than a Sharpie.
Head tilt is not quite far back enough, but ... it looks better rendered in something less blunt than a Sharpie.
Drawing every day.
-the Centaur
P.S. And 2000+ words, only ~3000 left for the month. Go nano!
Quick Sharpie sketch of Arnold Schwarzenegger, roughed first in non-repro blue as I really wanted to work on the eyes and landscape first, then cleaned up and very lightly shaded in Photoshop as it's late and I want to turn in. Eh, the eyes are still a notch too big, and lopsided, despite my efforts, but, it's not as completely terrible as some of the other sketches.
Drawing every day.
-the Centaur
P.S. and 2400+ words too.
Suuuper quick Sharpie sketch of a character from Stargate. Not terrible, but I think I need to focus on getting that egg shape of the human head a little bit more consistent (and one of the eyes got squished, but then, it's a suuuper quick Sharpie sketch without any roughs).
Drawing every day.
-the Centaur
P.S. Wrote 3600+ words today (and 8100+ words yesterday, when all was said and done).