Marrying classical CV and Deep Learning. I do things, which work, rather than being novel, but not working.
Czechia
- Tweets 22,309
- Following 795
- Followers 27,989
- Likes 79,506
Pinned Tweet
Update in the pinned post:
The best way to support Ukraine is via donating to the trusted funds. A few of them
@BackAndAlive - oldest and most trusted
savelife.in.ua/en/donate-en/…
@sternenko - father of drones
sternenkofund.org/en/donate
Czech one: @DarPutinovi
zbraneproukrajinu.cz
1
19
11,810
Dmytro Mishkin retweeted
We're nothing but reality tokenizers for our LLM overlords (we're data annotators)
1
1
10
759
Dmytro Mishkin retweeted
📢📢📢 introducing 𝐑𝐨𝐛𝐨𝐂𝐥𝐨𝐭𝐡
Neural materials still learn from 100 BRDFs measured in 2003 at MERL. As of today, that is no longer true (for cloth materials).
home: theialab.github.io/robocloth
arXiv: arxiv.org/pdf/2610.21090
500 cloths × 580 views – 80 billion observations
2
11
105
3,796
Dmytro Mishkin retweeted
This is a really important post, for a few reasons.
First, people need to know the scale of the problem and hear some radical (?) proposals on how to stop it.
Second: it is crucial to signal to bad actors, don't waste your time trying. It won't benefit you.
Some thoughts on reviewing: spylab.ai/blog/reviewing/
First for @satml_conf where we trialed an early reject phase which cut submissions by ~40%.
Then more generally about how not everything deserves peer-review, and why we should reward writing things that people actually read.
3
9
127
13,592
Dmytro Mishkin retweeted
Our new #rust kornia-slam pipeline is finally ou.
Check the site for more details: kornia.org/slam/
It’s fully #opensource — we are looking for battle testing users and contributors to improve it. Feel free to reach out to us !
Rust製のオープンソースSLAM 【kornia-slam】
🔹単眼・ステレオカメラやIMUを使い、カメラの位置推定と周囲の3D地図作成に対応
🔹再訪した場所を認識して、移動に伴う位置推定の誤差を補正
🔹WebカメラやOAK-Dにも対応。地図や推定結果を画面上で確認可能
Korniaチームが開発。
情報元はリプ欄。
4
30
3,414
Accepted to #WACV2027
DF26: We Cannot Tell Fake From Real Anymore
Severyn Shykula, Andrii Yermakov, Ivan Samarskyi, @ducha_aiki Jan Cech @Ana_Geneva
tl;dr: "public speaking" dataset, with semi-recent vidgen. Humans cannot reliably detect fakes.
P.S. No sound, silent video only
arxiv.org/abs/2609.07369
huggingface.co/datasets/DF26…
1
3
37
3,438
Dmytro Mishkin retweeted
Rust製のオープンソースSLAM 【kornia-slam】
🔹単眼・ステレオカメラやIMUを使い、カメラの位置推定と周囲の3D地図作成に対応
🔹再訪した場所を認識して、移動に伴う位置推定の誤差を補正
🔹WebカメラやOAK-Dにも対応。地図や推定結果を画面上で確認可能
Korniaチームが開発。
情報元はリプ欄。
3
19
269
12,743
Dmytro Mishkin retweeted
Linus Torvalds says, rules prohibiting AI usage are “UTTERLY STUPID” and “idiotic”.
For the Linux kernel, Torvalds says, “we’re not going to make stupid rules that say ‘you can’t use tools for X’”.
163
818
7,405
493,895
Dmytro Mishkin retweeted
❗️Last night, the operation of Starlink terminals in Ukraine was restricted.
Connectivity began to drop when moving at speeds exceeding 160 km/h. As a result, a number of critical missions were disrupted.
SpaceX likely took these measures at the request of the Trump administration. Also last night, when Kyiv was attacked by ballistic missiles, Ukraine did not receive intelligence regarding their launch— a failure that could have led to higher casualties.
Ukraine did not attack Russian oil refineries, yet Trump has already begun imposing sanctions on my country. Meanwhile, sanctions against Russia have been eased.
1,131
6,251
23,244
1,316,760
Dmytro Mishkin retweeted
Astra dethroned
Gemini 4 Argon is the best "vision" model
#1 for detection:
- tight, precise boxes
- dense scenes
#1 for text extraction:
- small text
- handwritten text
- complex tables
- rotated text
4× cheaper than Astra at high
The moment you've been waiting 4…
Our new frontier model, Gemini 4 Argon, is rolling out now in the Gemini app to bring your most complex ideas into reality.
See how Gemini 4 Argon and our latest updates across the Gemini app help you get the most out of every prompt. 🧵
35
81
1,182
127,360
To have accuracy as "circle size" now in my top of misleading plot examples. Not to mention, that star doesn't obey even this strange idea
2
1
11
1,639
Dmytro Mishkin retweeted
What's the best way to donate directly to Ukraine's efforts to destroy Russia's oil refineries and Diesel capacity?
Please send links or DM 🙏
10
4
64
7,777
Dmytro Mishkin retweeted
The amount of parallels between agent wrangling and student advising is enormous 😊
PhD Advising vs Vibe Coding #1:
One of the key skills a PhD advisor learns is context switching. Students come into your office every 30-60 minutes and you need to remember their project, where you left off, and what they were supposed to do next. It's a learned skill and exactly what you need to manage AI agents.
2
16
3,263
Just finished “Doom guy” by @romero. Great and honest book, one of the best in recent years.
Thank you for my childhood and inspiration now.
2
5
947
Agents finally brought the promise of all data labeling companies failed to deliver - you can label a little of data, new model/alg is trained on them and helps you more and more.
The labeling is becoming adaptive as well. I love this
5
16
2,500
Dmytro Mishkin retweeted
Yes. HW/SW is optimized for speed, not for determinism.
Main issue is that floating point operations result depend on the order in which they are computed, even if they are associative.
(a + b) + c can be different from a + (b + c) on a computer because of finite precision used.
As soon as you have some parallelism you cannot guarantee order, and your determinism is gone.
From a higher point of view, this leads to several reasons for non determinism, see nitter.siddz.com/AIQuanting/status/2107…
Fair. Temp 0 removes sampling, and everything else that makes inference nondeterministic stays. Batch shape, kernel reduction order and the GPU you land on all move the logits.
1
1
16
1,535
Fuck yes
It sounds insane, because we're used to a particular way of doing research: of giving and receiving credit, of counting beans, of hiring, recruiting, and promoting. And the machine god came to crush that game. The old game is dead. That reasonably causes incredible amounts of dread for many folks, who understand this as the death of research.
But it's not about Research, but the way we played the game. The old game was built around scarcity, scarce compute, scarce students, scarce hours, scarce people who knew the right technique and method etc. Most of a research career went into managing and playing along these constraints. Those walls just came crushing down. One person with a good question can now do what used to take a lab and five years.
You do not need permission, you do not need a big grant, you do not need a professor or researcher title, or a team of twenty experts , or one thousand gpus to stare straight into the first light of the universe.
How absolutely incredible
1
20
5,537
The best and the worst work for the agents is speed optimizations. Best - because it is well defined and easy to climb leaderboard.
Worst - because you cannot use agent swarm, as the timings would be unreliable.
3
1,033