Post
Conversation
![David Watson 🥑](data:image/svg+xml,<svg xmlns="http://www.w3.org/2000/svg" width="96" height="96"><rect fill-opacity="0"/></svg>)
Post your reply
I've tested it. The 3D Body works well. All the other 3D models are extremely low quality.
Have you had a chance to play with it? I’m always skeptical about 3D model generations. Maybe for simple models. I’ll have to check it out though. Thanks 
These examples seem suspiciously good. I'd like to see a real example from someone who has tried it
Can't wait to make models of models, and print these models for a model collection of models on a shelf full of models
I tested it briefly in the playground. I went in thinking that the 3D geometey reconstruction looks near perfect in the their demo video. And what about the textures? Then I got low quality results in both. My source was a photo-real rendering though, not a real photo.
This feels like a Trojan horse of sorts. Everyone else is trying to build a brain while Meta is focusing on the eyes, hands, and environment.
Clear applications here for creators, AR/VR, and eventual consumer hardware.
Well first you start with thousands of images, then select the few that work well, then you add animation and transitions to make everything move fast so your brain can't detect the mistakes in the models. Last you add some music. Pretty basic video editing if you ask me.
Wondering why they show some objects 360 degrees, whereas some only like 90 degrees
The Rootly team is heading to AWS re:Invent, and we’re bringing two Happy Hours designed for engineering leaders to connect, trade ideas, and unwind in great venues.
Monday Dec 01: Happy Hour on a cozy terrace with , , , , and .
Hard to believe we’re at the point where a picture instantly becomes a 3D model. This is a different era.
Would be so cool to be able to then hit “print” and your 3D printer goes to work.
Wow game dev potentially got a lot easier for creating nice environments
They have well-trained silhouette data to speed the render and fill the gap while AI does ML at the same time. I remember TRY doing this in 2005 with CV early development for a project.
So you could take photos and then use the 3d models for making a game? What are other cool things to use this for?
I am going to play with it, it could be a gamechanger for my business.
I wonder if you could select 4 or 5 still frames from a video to build a highly accurate 3D model for AI video generation? Can you imagine the special effects and scene setting control?
Wow 
This is the end of some CG artists and just the beginning for others…
Very cool though, especially for large scale shots in Hollywood films
This is the end of some CG artists and just the beginning for others…
Very cool though, especially for large scale shots in Hollywood films
From uncovering global vaccine injury to testifying in Senate hearings and standing in courtrooms across the country, the McCullough Foundation shows up where others won’t — confronting captured institutions, protecting the injured, exposing false claims, and fighting for ethical
1:45
The rigging part looks quite interesting. Are the rigs good enough for the humanoids?
Bro, the NSA did this shit to Will Smith back in the day. This isn't new tech.
Hamas keeps choosing terror. For freedom, Hamas must go.
Wow! Almost there to analyze and coach something as complicated as foil fencing. Looking forward to when video input drops!
SAM 3D just dropped the fattest nuke on our entire industry 
Photogrammetry studios → unemployed
Stock 3D furniture libraries → obsolete
Real estate agents can now pull perfect 3D pieces from any listing photo. Interior designers get infinite custom furniture on demand.
Photogrammetry studios → unemployed
Stock 3D furniture libraries → obsolete
Real estate agents can now pull perfect 3D pieces from any listing photo. Interior designers get infinite custom furniture on demand.
tried it out, the topology is nowhere need as clean as what they show in this video
Cool. Does anyone know who to export the 3d models?
Would be cool to import into AutoCAD for work
Cheers
Its great until the edge case where the back actually is not congruent with the front
It looks like it's just using image recognition to grab premade models of the first thing it recognizes and has a model for.
Your Personal AI, darlin™, has arrived on PC!
A 3D anime AI girl who brightens your everyday.
Customize her appearance, voice, and personality just the way you like — even upload your favorite models!
Download for Windows. Pair your smartphone too!
Still clumsy and making s lot of mistakes. Looks nice in commercials.
Can someone tell me what is the use case of this 3D models extracted from the images??
Ha, incredible. I will try it immediately. Just find that kind of app for 3 days and try a lot. I work for hand drawing —》 illustrator dimensions defined go for Gemini/chatgpt do rendering send image to AI app build 3D …so for now haha
Oh this good!
We’ll need this going forward to create VR spaces WITH objects.
I keep thinking of Sherlock Holmes’ flat at 221B. Walk into that in VR and be able to pick up anything and see what case he’s working on or experiment he’s in the middle of, just sit down and read
.
.
that means i can use this model's result to print anything i like with 3d printer?
I even granted accses but thing is blowed up to my face with pytorch error
Hilarious
Perfect for Christmas
Sale ends soon—don’t miss it. Shop Now!
Perfect for Christmas
Sale ends soon—don’t miss it. Shop Now!
Love the edit and scenes. Kind of odd the "toro" aka forklift in barcelona park güell from gaudí.
can we use this to create a 3D model and import to blender software?
Wow, SAM 3D! Seriously, who would have imagined? Meta keeps pushing tech boundaries. Can't wait to try creating jaw-dropping 3D art-almost feels like magic, doesn't it?
It's convenient, but I guess for high quality projects, one needs the option to add more angles for precision. Control is key
The Millermatic® 211 PRO powers through with USB-enabled upgrades and extended arc time. Built to crush every project, start to finish.
i guess none of the labs have ever thought that meta will come up with something like this
I'd be surprised if image scanning technology doesn't jump leaps and bounds from this.
I’m obsessed with people who are obsessed. This is what true product obsession sounds like:
This is wild — turning any still image into a clean 3D model changes the workflow entirely. Curious to see what creators can build with SAM 3D 
this is gona change game dev workflows completly, imagin just snappin reference objects
I hate it when companies I hate produce cool stuff I like. Makes me conflicted.
it is at the stage of meshy.ai, you can do the same thing..still not at a professional level, you would have to go in and fix polys and fine tune rigging and and all that..by the looks of what it shows it just has a catalogue of world items in that vid..i'd like to
It's an amazing start, and should speed up work that relies on 3D modeling/animation
Cool and all, but why the hell did removed image layers sam anything? I don't want to apply filters, I want to extract pngs -.-
The human body is a template. As for everything else — they lie to us, it's far from good.
its not working as accurate as the video , are they having a secret model which is not released ?
This Post is unavailable. Learn more
No one tells you that once you find the right routine for your neurospicy brain, you'll cry.
Because life WAS harder.
You can finally focus and do things easily without getting overwhelmed.
If you're struggling, try The Fabulous.
