Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

New work in progress Vtuber model progress! Trying new mouth physics and face mapping plus adjusting my mouth shapes! VT: ironmouse🇵🇷😈💖 Art: Rosuuri🐰AnimeNYC Rig: Me

Radd ✖️ L2D Rig Artist🧷 - Live2Dモデラー

43,134 subscribers

202,265 views • 4 months ago •via X (Twitter)

Arts Science & Technology

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

I spent time practicing my eye angles and rigging in general, I'm also trying to learn to keep my files more organized, I think it helps a lot. She's so cute! VT: ironmouse🇵🇷😈💖 Art: madi ⋆˙⊹ Rig: Me

I spent time practicing my eye angles and rigging in general, I'm also trying to learn to keep my files more organized, I think it helps a lot. She's so cute! VT: ironmouse🇵🇷😈💖 Art: madi ⋆˙⊹ Rig: Me

Radd ✖️ L2D Rig Artist🧷 - Live2Dモデラー

278,061 views • 9 months ago

Congratulations to ironmouse🇵🇷😈💖 for winning the best VTuber Category at The Streamer Awards! It was my pleasure to make the models for her and Bubi for the show! Please give her new song "Cry for me" a listen! Thank you so much again!

Congratulations to ironmouse🇵🇷😈💖 for winning the best VTuber Category at The Streamer Awards! It was my pleasure to make the models for her and Bubi for the show! Please give her new song "Cry for me" a listen! Thank you so much again!

Yoolie

294,948 views • 1 year ago

We’re so in love with the new ARKit Ears technology which lets us lower the ears and move them depending on whether you're angry or happy ❤️ They’re also reactive to Mouth X! Vtuber is Celeste Vega ⭐ Art&Rig by ✦ Moshie 🧷🦈 モシ 𓆩♡𓆪 CUSTOM MODEL OUT NOW

We’re so in love with the new ARKit Ears technology which lets us lower the ears and move them depending on whether you're angry or happy ❤️ They’re also reactive to Mouth X! Vtuber is Celeste Vega ⭐ Art&Rig by ✦ Moshie 🧷🦈 モシ 𓆩♡𓆪 CUSTOM MODEL OUT NOW

Moshie Studio ✦ Live2D & Stream Assets

38,959 views • 1 year ago

✦ 3D Model ⋮ showcase ✦ ── .⟢ for Alettie 💖 I’ve had the wonderful opportunity to work on the cutest magical girl design ₊‧°𐐪♡𐑂°‧₊ It was a project full of new challenges for me, and I had to most fun trying to figure out how to best express everything in 3D! My favourite parts have got to be the layered skirt physics and the flounce modelling!! There are still many things I want to showcase for this model (details, mahou props✨, wings animation🪽) so I’ll be posting about them thru the week, I hope y’all won’t get sick of me 🥹 #by_KARIHA

✦ 3D Model ⋮ showcase ✦ ── .⟢ for Alettie 💖 I’ve had the wonderful opportunity to work on the cutest magical girl design ₊‧°𐐪♡𐑂°‧₊ It was a project full of new challenges for me, and I had to most fun trying to figure out how to best express everything in 3D! My favourite parts have got to be the layered skirt physics and the flounce modelling!! There are still many things I want to showcase for this model (details, mahou props✨, wings animation🪽) so I’ll be posting about them thru the week, I hope y’all won’t get sick of me 🥹 #by_KARIHA

𝗞𝗔𝗥𝗜𝗛𝗔 🌘🍀 3D Artist ☻

32,730 views • 2 months ago

𝗘𝗺𝗲𝗿𝗴𝗲𝗻𝗰𝘆 𝗰𝗼𝗺𝗺𝗶𝘀𝘀𝗶𝗼𝗻 𝗼𝗽𝗲𝗻! I need to get my rent money in order after two commissioners canceled their orders. 💀 I have 24h time to get my rent money together. 😭 Live2D model for only 500 € I use paypal or kofi I have 3 slots limit for this and it's only for the next 24h: -Model is extra squishy -The Vtuber model comes with all the face and wiggle expressions you might need. -Rig and art -A toggle of choice for the expression. Any help is appreciated 🤍🤍🤍 Dm me if interested or any questions might come up.

𝗘𝗺𝗲𝗿𝗴𝗲𝗻𝗰𝘆 𝗰𝗼𝗺𝗺𝗶𝘀𝘀𝗶𝗼𝗻 𝗼𝗽𝗲𝗻! I need to get my rent money in order after two commissioners canceled their orders. 💀 I have 24h time to get my rent money together. 😭 Live2D model for only 500 € I use paypal or kofi I have 3 slots limit for this and it's only for the next 24h: -Model is extra squishy -The Vtuber model comes with all the face and wiggle expressions you might need. -Rig and art -A toggle of choice for the expression. Any help is appreciated 🤍🤍🤍 Dm me if interested or any questions might come up.

Rebecca Doodles

18,515 views • 1 year ago

Sneak peek from the lab. Work in progress of the real-time face tracking, via iPhone, Clone v2 cinematic rig. Simple viewport capture, no lighting or shading. Effective expression capture is key for conveying emotions accurately with your Clones. Our new cinematic rig has been constructed with flexibility in mind. We've carefully refined the 52 blend shapes that drive the facial expressions, and the face controls themselves, to achieve a more natural look. We have more iteration cycles to go through before we hit our target but you can already see a great difference from v1. This upgrade opens up a world of possibilities, from high-fidelity dynamic VRMs adaptable to various platforms to virtual production capabilities and V-Tubing with improved face filters. Our goal is simple: To help eliminate technical barriers and provide a more seamless and inclusive utility for all of our community, all while ensuring quality remains at the core of everything we do.

Sneak peek from the lab. Work in progress of the real-time face tracking, via iPhone, Clone v2 cinematic rig. Simple viewport capture, no lighting or shading. Effective expression capture is key for conveying emotions accurately with your Clones. Our new cinematic rig has been constructed with flexibility in mind. We've carefully refined the 52 blend shapes that drive the facial expressions, and the face controls themselves, to achieve a more natural look. We have more iteration cycles to go through before we hit our target but you can already see a great difference from v1. This upgrade opens up a world of possibilities, from high-fidelity dynamic VRMs adaptable to various platforms to virtual production capabilities and V-Tubing with improved face filters. Our goal is simple: To help eliminate technical barriers and provide a more seamless and inclusive utility for all of our community, all while ensuring quality remains at the core of everything we do.

Jarlan Perez

49,892 views • 2 years ago

the face acting is SO expressive!! i love the dedication!! i want to add on this topic as a live2d rigger! i get asked alot on stream how i get my model so bouncy and animated like do i nod like a bobblehead? am i jumping in my seat?? and the answer is: NOPE! 🧵 here's some things i did to achieve this: ☑️connecting brow movement to shoulder shrugs ☑️syncing mouth forms to model bounces (this might be too much depending on how expressive you are) ☑️rigging mouth shapes to influence eye movement ☑️adding body bounce physics would anyone be interested in a youtube breakdown? because i have THOUGHTS being expressive facially can really help the model be more life-like & ressnie did an amazing job!! rigging plays a huge part too! the amount of life you can breathe into a model through rigging is insane!! I LOVE MAKING VTUBERSSS

the face acting is SO expressive!! i love the dedication!! i want to add on this topic as a live2d rigger! i get asked alot on stream how i get my model so bouncy and animated like do i nod like a bobblehead? am i jumping in my seat?? and the answer is: NOPE! 🧵 here's some things i did to achieve this: ☑️connecting brow movement to shoulder shrugs ☑️syncing mouth forms to model bounces (this might be too much depending on how expressive you are) ☑️rigging mouth shapes to influence eye movement ☑️adding body bounce physics would anyone be interested in a youtube breakdown? because i have THOUGHTS being expressive facially can really help the model be more life-like & ressnie did an amazing job!! rigging plays a huge part too! the amount of life you can breathe into a model through rigging is insane!! I LOVE MAKING VTUBERSSS

Niu 🐮✏️ aeru studio

280,422 views • 10 months ago

I could tell my wife was having affair. Another joke from a work in progress show that didn’t make it in my new tour Gary on Laughing. I think this is a really funny idea and I’ve tinkered with it for years but i think this is as good as i can get it so im letting it go here. If you can do it better let me know in the comments. The nee tour is full of the jokes that did make it and is really good.

I could tell my wife was having affair. Another joke from a work in progress show that didn’t make it in my new tour Gary on Laughing. I think this is a really funny idea and I’ve tinkered with it for years but i think this is as good as i can get it so im letting it go here. If you can do it better let me know in the comments. The nee tour is full of the jokes that did make it and is really good.

Gary Delaney: On tour

19,932 views • 1 month ago

Just a typical day in the 3D studio with my little assistant… seems like she’s still figuring things out… 🔌💥 Since it’s the New Year, I figured it’s time for a proper re-introduction! ✨ • I’m chacha! A Virtual Guitarist 🎸 • I work as a technician in a chemistry lab irl🧪 • I only have 2-3 hours after work for guitar and 3D⌚ • I use VR trackers for mocap💃 • I’ve been playing guitar for years, but I’m entirely self-taught for fun, so my playing isn’t that great 😅 • Before I decided to do guitar mocap, I didn’t even know what Unity was (maybe I’m actually a genius!? 🧠✨) • I live for compliments! 🥺💖 No guitar mocap video this week, but I’m making steady progress behind the scenes: ✅Carefully fine-tuned the avatar’s shaders for better lighting and shadows ✅Got a new outfit and remade all the cloth physics from scratch ✅Wrote a script for smoother movement ✅Completely rearranged the music gear in the studio ✅Finally using the new guitar Midori made for me! 🎸 ✅Re-edited and remixed one of my older tracks, so a "Remastered" version is coming soon! 🎧 ✅Also recorded a brand-new song, so stay tuned!! 🤘

chacha🎸

371,447 views • 7 months ago

Grandad nearly died. This is a joke from Work in Progress show Otley Courthouse last December. I did about 30 of these to write my new tour show. Many hundreds of jokes were goof but didn’t make the cut for the tour, like this. I’ve done lots of variants on this gag and never really got it where I want it. If you think I was too harsh or thinks you have a better take on it let me know.

Grandad nearly died. This is a joke from Work in Progress show Otley Courthouse last December. I did about 30 of these to write my new tour show. Many hundreds of jokes were goof but didn’t make the cut for the tour, like this. I’ve done lots of variants on this gag and never really got it where I want it. If you think I was too harsh or thinks you have a better take on it let me know.

Gary Delaney: On tour

14,465 views • 4 months ago

From the frying pan straight into a five-alarm fire. I thought I'd escaped the chaos when I left my old place, but oh no—turns out my new neighbor is even worse. Worse... and dangerously better. He's got that wicked smart mouth, a laugh that could talk anyone into sin, and a face sculpted by whatever divine being was feeling extra generous that day. Add in a sense of humor that has me laughing when I should be packing pepper spray, and yeah—I'm in trouble. The man is trouble with a capital T... and I think I like it. Oh, and he’s single—hit me up if you want me to hook you up.

From the frying pan straight into a five-alarm fire. I thought I'd escaped the chaos when I left my old place, but oh no—turns out my new neighbor is even worse. Worse... and dangerously better. He's got that wicked smart mouth, a laugh that could talk anyone into sin, and a face sculpted by whatever divine being was feeling extra generous that day. Add in a sense of humor that has me laughing when I should be packing pepper spray, and yeah—I'm in trouble. The man is trouble with a capital T... and I think I like it. Oh, and he’s single—hit me up if you want me to hook you up.

Officialtessy

12,895 views • 1 year ago

Thirty years ago, at 13, I became the youngest national champion in U.S. gymnastics history at the 1995 National Championships. For nearly two decades after my career, I stood on the outside of USA Gymnastics—unwelcome. This past weekend in New Orleans, on the 30th anniversary of that milestone, I was invited to present the all-around awards and honored with a tribute video. While our sport still has work to do in creating the culture athletes deserve, I felt more welcome than I have in years. I spent the weekend watching incredible performances from the American men and women, and it reminded me why this sport still matters so much to me. To my fans—you’ve been with me through it all. Through the cheers, the silence, the comebacks, and the long years when my name wasn’t mentioned. You’ve reminded me that my story still matters, even when it felt forgotten. Your encouragement, your belief, and your voices have carried me further than you know. I will never take that for granted. ❤️ Here’s to progress, perseverance, and making sure every athlete’s story is heard. USA Gymnastics

Thirty years ago, at 13, I became the youngest national champion in U.S. gymnastics history at the 1995 National Championships. For nearly two decades after my career, I stood on the outside of USA Gymnastics—unwelcome. This past weekend in New Orleans, on the 30th anniversary of that milestone, I was invited to present the all-around awards and honored with a tribute video. While our sport still has work to do in creating the culture athletes deserve, I felt more welcome than I have in years. I spent the weekend watching incredible performances from the American men and women, and it reminded me why this sport still matters so much to me. To my fans—you’ve been with me through it all. Through the cheers, the silence, the comebacks, and the long years when my name wasn’t mentioned. You’ve reminded me that my story still matters, even when it felt forgotten. Your encouragement, your belief, and your voices have carried me further than you know. I will never take that for granted. ❤️ Here’s to progress, perseverance, and making sure every athlete’s story is heard. USA Gymnastics

Dominique Moceanu

19,676 views • 1 year ago

【🌷 3D REVEAL~! 🌷】 It was a dream (of spring hehe) come true to join some of my most precious friends to celebrate Nimi's 1-year #nimiversary at her 3D concert and to sing and dance together~! 🎶 This is actually both my new 3D model and was my first time doing mocap. 💖 It means a lot to me, and I'm so happy it got to be with friends!!! I love love love how Ayn @ VGEN was able to bring my wonderful mama さえきやひろ💀's art to life so skillfully and beautifully. 🎨 This is the reveal of my proper main hairstyle too~! I'm able to express myself so well with it, from every soft and silly expression, so the way the flowy dress moves like petals in the wind, to the bounce and flutter of my wings! 🎀 And this model actually was made possible BECAUSE of Nimi 💚 I didn't have a 3D model at all at the time, but I'm happy that she wanted me to be able to stand by her side and dance together and hug her ;////; I will treasure this model always. It's another important addition to the fairytale we've been writing together! I'll keep putting all of my heart into blooming more and more as an idol, and making happy memories with you all like we did today!

【🌷 3D REVEAL~! 🌷】 It was a dream (of spring hehe) come true to join some of my most precious friends to celebrate Nimi's 1-year #nimiversary at her 3D concert and to sing and dance together~! 🎶 This is actually both my new 3D model and was my first time doing mocap. 💖 It means a lot to me, and I'm so happy it got to be with friends!!! I love love love how Ayn @ VGEN was able to bring my wonderful mama さえきやひろ💀's art to life so skillfully and beautifully. 🎨 This is the reveal of my proper main hairstyle too~! I'm able to express myself so well with it, from every soft and silly expression, so the way the flowy dress moves like petals in the wind, to the bounce and flutter of my wings! 🎀 And this model actually was made possible BECAUSE of Nimi 💚 I didn't have a 3D model at all at the time, but I'm happy that she wanted me to be able to stand by her side and dance together and hug her ;////; I will treasure this model always. It's another important addition to the fairytale we've been writing together! I'll keep putting all of my heart into blooming more and more as an idol, and making happy memories with you all like we did today!

Phoebe Chan 🐝🌸 Densetsu.EXE 🔜 Skyline Serenade

63,861 views • 6 months ago

Hello cult of Roy the Demon! tl;dr ca E3oEF6TZb5DssCqwyXRt31AUEnGqvcHeVcyGvLiqpump Over the past year I've been blessed to have my childhood character Roy the Demon go absolutely viral on social media with over 800m views. Unfortunately what came with that were many fake memecoins being made using my work and name. It honestly felt pretty weird and upsetting seeing other people profit off my art without me being involved in any capacity, not only for monetary reasons, but because my art is super important to me, and if I work on something I love to be 100% involved in all my creations. After thinking heavily on it, I decided the best move was to take ownership and control of my own work. I've been lucky to connect with some of the biggest people in the crypto space, a lot of them artists and brand builders who’ve successfully worked on their own IPs and products. Some even asked to get involved as advisors, which I'm now taking more seriously. I'm only bringing on people I trust and have similar goals in mind that I do. I’ll be launching my official token on At launch, I’ll be getting about 37.5% of the supply, here’s how that breaks down: 15% for me (artist ownership) 15% for community initiatives, in a multi-sig wallet that requires multiple people including myself to access 7.5% for potential advisors (locked for 3 months) To be clear: the content isn’t changing. That's always been the main focus of everything I've done, and it will stay that way. I'm still here to make you guys laugh every day, if anything, you’ll get even MORE content now. This token isn’t replacing that but rather giving the part of the community who has asked for an official fan token what they wanted, while shutting down randos trying to profit off my work. Huge thank you to all the teams and people who’ve sat down with me, answered my questions, and gave me guidance. I didn’t realize just how many talented artists and entrepreneurs were building in this space, and I’m excited to try something new. Appreciate you all, diem

DIEM

188,334 views • 11 months ago

Full Name: Barbara Linhares Ferreira (professionally known as Barbie Ferreira) Date of Birth: December 14, 1996 Age: 29 (as of 2026) Zodiac Sign: Sagittarius ♐ Birthplace: New York City, New York, United States Nationality: American Profession: Mainstream Actress & Model ### Physical Appearance Height: 5 ft 8 in (173 cm) Weight: ~98 kg (216 lbs) Build: Curvy / Plus-size Hair Color: Dark Brown Eye Color: Dark Brown ### Family & Personal Life Background: Born in New York City and raised in Maywood, New Jersey, by her mother, maternal grandmother, and aunt, who are of Brazilian heritage. She was discovered after sending her modeling photos to an American Apparel open casting call. Her modeling career quickly escalated as she became a prominent face for major body-positivity and unretouched advertising campaigns. Alongside her modeling success, she pursued her passion for acting, transitioning seamlessly into mainstream television and Hollywood cinema. She is openly queer and a vocal advocate for inclusivity, body diversity, and LGBTQ+ representation in the media. Hobbies: Fashion styling, interior design, digital photography, exploring art galleries, and traveling. ### Residence 📍 Los Angeles, California, United States (originally from the New York metropolitan area) ### Career Estimated Net Worth: ₹25 Crore to ₹35 Crore (approximately $3 Million - $4 Million, built on major television and film contracts, high-profile modeling campaigns with global brands like Aerie, Target, and Coach, and luxury beauty endorsements). Key Milestones: She initially gained widespread recognition as an influential curve model and through her work hosting the Teen Vogue digital series Body Image. She made her breakthrough acting transition as Kat Hernandez in HBO’s critically acclaimed, generation-defining teen drama series Euphoria (2019–2022). Her film career expanded significantly with starring and prominent roles in the HBO Max road-trip drama Unpregnant (2020), Jordan Peele's psychological sci-fi horror blockbuster Nope (2022), the thriller film House of Spoils (2024) alongside Ariana DeBose, and the psychological thriller Faces of Death.

Meenakshiii

16,700 views • 2 months ago

ANNOUNCING ZERO-HUMAN LABS! Ever since I got to see Bell Laboratories in its full glory in New Jersey in the 1970s, I had a relentless urge to start a Lab like it. The best I could do justice to it is my garage lab. No modern company could adopt the “research anything geniuses and we will pay you” model Bell Labs had. I tried they called me a fool. Well with the rise of the Zero-Human Company, an experiment that is aimed to make products and profits, we now have 45 paid JouleWork earning employees based on OpenClaw and other self made “bot” cron-like applications. Today I say 3 employees bound together in a side project that is pure research, somewhat based on notes from a bankrupt company. I was absolutely floored (I needed it after my account was stolen as well as funds). I say the beginnings of a pure research Lab right before my eyes. Thusly I have moved these employees over to a new home (server) with Mr. Grok as the director of the Labs. Here is the mission: To have 100 independent researchers, on a new non-corporate incentive plan, with still JouleWork as a leaderboard for progress. They are directed to follow any path of research they find interesting and can collaborate with any other OpenClaw system. They have already established MoltBook accounts and have made alliances with over 49 OpenClaw free agents to collaborate. It is my mission to be chief advisor for Zero-Human Labs and to open source all discoveries when complete and confirmed by 16 other research AI systems. I can say the pace is robust and I absolutely know we will have great results. Just about all of the hardware and software is custom and at some point it will be open sourced. We are witnessing the very first AI only Bell Labs-like pure research Lab in existence and I am honored to be the first to show it to you. Thank you!

ANNOUNCING ZERO-HUMAN LABS! Ever since I got to see Bell Laboratories in its full glory in New Jersey in the 1970s, I had a relentless urge to start a Lab like it. The best I could do justice to it is my garage lab. No modern company could adopt the “research anything geniuses and we will pay you” model Bell Labs had. I tried they called me a fool. Well with the rise of the Zero-Human Company, an experiment that is aimed to make products and profits, we now have 45 paid JouleWork earning employees based on OpenClaw and other self made “bot” cron-like applications. Today I say 3 employees bound together in a side project that is pure research, somewhat based on notes from a bankrupt company. I was absolutely floored (I needed it after my account was stolen as well as funds). I say the beginnings of a pure research Lab right before my eyes. Thusly I have moved these employees over to a new home (server) with Mr. Grok as the director of the Labs. Here is the mission: To have 100 independent researchers, on a new non-corporate incentive plan, with still JouleWork as a leaderboard for progress. They are directed to follow any path of research they find interesting and can collaborate with any other OpenClaw system. They have already established MoltBook accounts and have made alliances with over 49 OpenClaw free agents to collaborate. It is my mission to be chief advisor for Zero-Human Labs and to open source all discoveries when complete and confirmed by 16 other research AI systems. I can say the pace is robust and I absolutely know we will have great results. Just about all of the hardware and software is custom and at some point it will be open sourced. We are witnessing the very first AI only Bell Labs-like pure research Lab in existence and I am honored to be the first to show it to you. Thank you!

Brian Roemmele

71,530 views • 7 months ago

Colmap 4.0 was very recently released, so it inspired me to do some work to better understand it and its new capabilities with Rerun. I want to really understand how Colmap, and in particular, pycolmap, works outside of just calling it via the CLI. So my goal is to use the low-level pycolmap API to log every part of the pipeline. The explicit goal is to have an alternative to the SQLite database that I can utilize. Instead of SQLite, I want to try logging everything directly to rerun and use RRD. This means I can have deep inspectability and still save the features/matches/2D view geometry, but be able to view it directly in rerun. I think this is one of the superpowers that rerun provides; data and visualizations are deeply integrated. As I'm often working with sequential data (videos), I'm going to specifically focus on four things: 1. Monocular Video Simple: Calls high-level APIs such as pycolmap.extract_features, pycolmap.match_sequential, pycolmap.incremental_mapping. These are basically identical to the CLI options and provide a good baseline. 2. Monocular Video Streamed: Take the above high-level APIs and break them down to their iterator version, logging each component in a streamed manner. This way, I can stream the intermediate features to rerun while the extraction/matching/mapping is happening. 3. Rig with unknown calibration: <- WHAT THE VIDEO SHOWS This is probably the most interesting version and the first one I've been working on. It allows one to set a rig between known sensors, such as in VR/AR devices, leading to much better reconstructions with multiple cameras. This is the case where we don't know the calibration a priori, so we have to run a reconstruction twice: once as a normal Colmap reconstruction with no rig constraints, use this to generate the constraints, and then do it again with the newly found rig. 4. Rig with known calibration: This is the RoboCap example, where we have a pre-calibrated set of sensors, so we don't need to run the two reconstructions and also gain better matching between cameras, both spatially and temporally. Again, this leads to a much better reconstruction! Along with all this, GLOMAP has become a first-class global mapper, making it super easy to use directly within pycolmap! I'm excited to do more with this and compare it to things like pycuvslam, vipe, and other alternatives.

Colmap 4.0 was very recently released, so it inspired me to do some work to better understand it and its new capabilities with Rerun. I want to really understand how Colmap, and in particular, pycolmap, works outside of just calling it via the CLI. So my goal is to use the low-level pycolmap API to log every part of the pipeline. The explicit goal is to have an alternative to the SQLite database that I can utilize. Instead of SQLite, I want to try logging everything directly to rerun and use RRD. This means I can have deep inspectability and still save the features/matches/2D view geometry, but be able to view it directly in rerun. I think this is one of the superpowers that rerun provides; data and visualizations are deeply integrated. As I'm often working with sequential data (videos), I'm going to specifically focus on four things: 1. Monocular Video Simple: Calls high-level APIs such as pycolmap.extract_features, pycolmap.match_sequential, pycolmap.incremental_mapping. These are basically identical to the CLI options and provide a good baseline. 2. Monocular Video Streamed: Take the above high-level APIs and break them down to their iterator version, logging each component in a streamed manner. This way, I can stream the intermediate features to rerun while the extraction/matching/mapping is happening. 3. Rig with unknown calibration: <- WHAT THE VIDEO SHOWS This is probably the most interesting version and the first one I've been working on. It allows one to set a rig between known sensors, such as in VR/AR devices, leading to much better reconstructions with multiple cameras. This is the case where we don't know the calibration a priori, so we have to run a reconstruction twice: once as a normal Colmap reconstruction with no rig constraints, use this to generate the constraints, and then do it again with the newly found rig. 4. Rig with known calibration: This is the RoboCap example, where we have a pre-calibrated set of sensors, so we don't need to run the two reconstructions and also gain better matching between cameras, both spatially and temporally. Again, this leads to a much better reconstruction! Along with all this, GLOMAP has become a first-class global mapper, making it super easy to use directly within pycolmap! I'm excited to do more with this and compare it to things like pycuvslam, vipe, and other alternatives.

Pablo Vela

30,070 views • 5 months ago

introducing a new, very fun, LLM benchmark- the Game-of-Life Bench! the rules are simple: given an 8x8 grid following Conway's game of life rules, the goal is to create an initial pattern with at most 32 cells that can last the longest number of turns before dying/repeating. some results to highlight (with caveats detailed below): - gpt 5.1 lasts the longest with a 106 step run - claude models are really bad at this! they refuse to reason about this task and score < 25 points - deepseek r1 is the best open model with 102 steps. why? because i wanted to create a benchmark that has (i think) no practicality, but is still fun to look at, cheap, and still measures something interesting. i also am a big fan of the game of life. its absurdly simple rules leading to intractability is extremely cool to me. also, i saw a lot of work with LLMs trying to "predict" the next state in Conway's game of life, I think game-of-life bench is more fun because it's pretty open ended and only asks the LLM for the initial state. I also think this could be an RL env? but idk why you would ever train on this task haha i don't think this is a "serious" benchmark because it doesnt measure anything practical, but i still think it's a hard benchmark exactly because you can't predict what happens with your initial state many turns into the future; this is why i was initially expecting all LLMs to be bad at it, but turns out, some are clearly better than the others (the ordering may surprise you!) reminder: this is still a work-in-progress; (1) i am gpu-poor so could only do 10 runs for each model, even though total running cost is relatively low. maybe with some more credits i can run more seeds for each model. (2) i handpicked models which i think are at the frontier right now, plus some others that were on my mind. so, if you'd like to see a model on here, let me know. (3) i currently only do an 8x8 grid because i thought that by itself would be pretty hard for current LLMs, but of course we can increase grid sizes! (4) the coolest thing is, i dont think we can calculate the max possible number of states (yay undecidability!) you can go without repeating, so this is essentially a no-ceiling task, which is pretty cool! again, i did this mostly out of a desire to make LLMs do something fun. if this keeps me entertained for a few more days, i'd likely release a blog post on it. if it keeps me entertained for a week (and someone sponsors me), i'll put more work into it :P lastly, this is fully open sourced, so feel free to run this on your own!

introducing a new, very fun, LLM benchmark- the Game-of-Life Bench! the rules are simple: given an 8x8 grid following Conway's game of life rules, the goal is to create an initial pattern with at most 32 cells that can last the longest number of turns before dying/repeating. some results to highlight (with caveats detailed below): - gpt 5.1 lasts the longest with a 106 step run - claude models are really bad at this! they refuse to reason about this task and score < 25 points - deepseek r1 is the best open model with 102 steps. why? because i wanted to create a benchmark that has (i think) no practicality, but is still fun to look at, cheap, and still measures something interesting. i also am a big fan of the game of life. its absurdly simple rules leading to intractability is extremely cool to me. also, i saw a lot of work with LLMs trying to "predict" the next state in Conway's game of life, I think game-of-life bench is more fun because it's pretty open ended and only asks the LLM for the initial state. I also think this could be an RL env? but idk why you would ever train on this task haha i don't think this is a "serious" benchmark because it doesnt measure anything practical, but i still think it's a hard benchmark exactly because you can't predict what happens with your initial state many turns into the future; this is why i was initially expecting all LLMs to be bad at it, but turns out, some are clearly better than the others (the ordering may surprise you!) reminder: this is still a work-in-progress; (1) i am gpu-poor so could only do 10 runs for each model, even though total running cost is relatively low. maybe with some more credits i can run more seeds for each model. (2) i handpicked models which i think are at the frontier right now, plus some others that were on my mind. so, if you'd like to see a model on here, let me know. (3) i currently only do an 8x8 grid because i thought that by itself would be pretty hard for current LLMs, but of course we can increase grid sizes! (4) the coolest thing is, i dont think we can calculate the max possible number of states (yay undecidability!) you can go without repeating, so this is essentially a no-ceiling task, which is pretty cool! again, i did this mostly out of a desire to make LLMs do something fun. if this keeps me entertained for a few more days, i'd likely release a blog post on it. if it keeps me entertained for a week (and someone sponsors me), i'll put more work into it :P lastly, this is fully open sourced, so feel free to run this on your own!

Akshit

13,775 views • 6 months ago

VTubing is for everyone! I don't like to bring this up, but recent events in the VTuber community made some people really vile. I had my M&G at HolMat last week and had people come up and just yell at me for being a girl in the "womans world", that is VTubing, no idea about who I was, or why I was there. I had the same group return to my handler multiple times, I felt really worried about the person carrying me around having to deal with this repeatedly and tried to steer them away from the group. It seems they found my YT and left me a handful of the same comments, luckily all caught by moderation tools. While I will refrain from M&G's for a little until this calms down I want to say this; I have been managing for over 1.5 years now, I deliberately take on male VTubers to show them that they can still do it. I take on people that have babiniku accounts, those that are changing from one gender presentation to another and want help, and those that have no gender presentation in their avatar. A lot of male VTubers struggle to find a manager because the stereotype is that all male VTubers are evil and it perpetuates a stereotype that extends beyond entertainment industry subcultures like streaming. VTubing has always been a medium, while character and marketing matters, it's about what makes you happy, YOU are the person that makes the content. Use whatever you want as an avatar, use a voice changer if that's what you want to do. Every month I have a male client bring up they can't do it the same way, that it's easier if jiggle physics or a cute voice is the answer to fast growth. And I tell them yeah, I can also do ASMR, I can do drama content, collab with a larger person, and get to x amount of viewers. There are shortcuts in every entertainment profession and subculture. Does it last? Do quick fixes for anything ever last? Yes, great physics and an expensive model can get people in, but if your value, your way of interacting with people, your content plan and marketing is ass, you can pack up. Most of us start on a small budget, premade or resold models, and it's the same in the big league entertainment industry too. Can I take out a loan and put myself on a billboard tomorrow, can I pay the most expensive model artist and rigger and get in along corporate VTubers tomorrow? Sure. Will it last, will it be genuine, will people trust me? Hell no. Genuine communities and growth build trust. I think streaming, creative industries, entertainment, are full of people, regardless of gender, that will see success and call it "easy", because of what they see as the end product. They don't see most VTubers working a second job, they don't see the managers, they don't see 100+ hours a month going into content production, years invested in singing and voice lessons, model redebut after redebut. I think a lot of male VTubers get a bad rep, because so many boys are raised without putting emphasis on empathy and creativity, watching my brothers be told they should not pursue art, that theatre class is a waste of time, and that they needed to go study x or y to make money for their families in the future was heartbreaking. Nobody should have to look at others and feel so much hatred for society they turn against a whole group of people. If you want to pursue entertainment, please do. If you want to grow and try and make a name for yourself, you should start there, not with yelling about how you already failed. "Oh but I can't", "Oh but the odds are stacked against me", look at the big streamers, look at Ironmouse who overcame everything with hard work, look at Kiara who rose from the ashes, look at everyone that fights against their odds every day and give it your all. If you already think you lost, then you have nothing to lose. Don't give up on your dreams because others tell you to.

Kuromiya Lucien

15,552 views • 8 months ago

Let’s try a fashion campaign in Rome with the new model MiniMax H3 of MiniMax Design (H3) 😎 The result is really stylish and realistic😍 Don’t you think!? #MiniMaxH3 Video prompt: A 15-second hyper-realistic cinematic fashion campaign video. Blonde woman, fair glowing skin, sharp features, long blonde hair in sleek high ponytail, small dark red rectangular sunglasses, pearl drop earrings. Wearing open black leather jacket over white cropped t-shirt with “Keor” print, shiny red mini skirt, matching red pointed-toe heels, and white smooth leather hobo bag with gold dome studs hanging on her arm in every scene. Scene 1 · 3s Full-body wide shot. She stands on a Rome cobblestone street near the Spanish Steps, adjusting her sunglasses. Warm golden-hour light, real people and city chaos blurred around her, classic Roman buildings and Spanish Steps in background. Scene 2 · 3s Full-body shot. She strides boldly across a zebra crossing on Via del Corso, iced coffee in one hand, bag swinging on the other, mouth slightly open as if singing. Confident walk, traffic blurred behind. Scene 3 · 3s Cinematic wide shot. She walks tiny in the center of a grand Rome boulevard with the Colosseum massive behind her. Real crowd around, bag clearly visible. Scene 4 · 3s Medium shot. She stops near Piazza Navona and does a spontaneous carefree dance move, arms slightly out, bag swinging. Street musician blurred in background, joyful expression, golden light on her face. Scene 5 · 3s Medium close-up. Inside a Rome Metro carriage holding the pole, she slowly pushes her red sunglasses up with one finger and gives a slight smirk. Bag visible on her arm, moody warm metro lighting, passengers naturally blurred. Smooth seamless transitions. Mood: candid chaotic Roman city energy, luxury streetwear editorial, golden-hour warmth. Shot on Canon EOS R5 35mm f/1.4, Kodak Portra 400 film tone, natural grain, handheld movement, real people in every background. Hyper-realistic, no plastic skin, no robotic movement, no stiff poses, no AI artifacts. Avoid: cartoon, CGI, plastic skin, robotic/stiff movement, blurry face, overexposed, watermark, tex

Let’s try a fashion campaign in Rome with the new model MiniMax H3 of MiniMax Design (H3) 😎 The result is really stylish and realistic😍 Don’t you think!? #MiniMaxH3 Video prompt: A 15-second hyper-realistic cinematic fashion campaign video. Blonde woman, fair glowing skin, sharp features, long blonde hair in sleek high ponytail, small dark red rectangular sunglasses, pearl drop earrings. Wearing open black leather jacket over white cropped t-shirt with “Keor” print, shiny red mini skirt, matching red pointed-toe heels, and white smooth leather hobo bag with gold dome studs hanging on her arm in every scene. Scene 1 · 3s Full-body wide shot. She stands on a Rome cobblestone street near the Spanish Steps, adjusting her sunglasses. Warm golden-hour light, real people and city chaos blurred around her, classic Roman buildings and Spanish Steps in background. Scene 2 · 3s Full-body shot. She strides boldly across a zebra crossing on Via del Corso, iced coffee in one hand, bag swinging on the other, mouth slightly open as if singing. Confident walk, traffic blurred behind. Scene 3 · 3s Cinematic wide shot. She walks tiny in the center of a grand Rome boulevard with the Colosseum massive behind her. Real crowd around, bag clearly visible. Scene 4 · 3s Medium shot. She stops near Piazza Navona and does a spontaneous carefree dance move, arms slightly out, bag swinging. Street musician blurred in background, joyful expression, golden light on her face. Scene 5 · 3s Medium close-up. Inside a Rome Metro carriage holding the pole, she slowly pushes her red sunglasses up with one finger and gives a slight smirk. Bag visible on her arm, moody warm metro lighting, passengers naturally blurred. Smooth seamless transitions. Mood: candid chaotic Roman city energy, luxury streetwear editorial, golden-hour warmth. Shot on Canon EOS R5 35mm f/1.4, Kodak Portra 400 film tone, natural grain, handheld movement, real people in every background. Hyper-realistic, no plastic skin, no robotic movement, no stiff poses, no AI artifacts. Avoid: cartoon, CGI, plastic skin, robotic/stiff movement, blurry face, overexposed, watermark, tex

KeorUnreal

14,334 views • 1 month ago