正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

8. CUSTOMIZABLE HEAD MODEL dont think many people know this unless they check csp updates (its more of a recent feature) but the default head model is completely customizable and it’s awesome!! can choose from presets or adjust features manually materials > 3D > basic

KITTY★DISK

8,572 subscribers

35,529 次观看 • 1 年前 •via X (Twitter)

科学技术教育艺术

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Our customizable 3D vtuber model now has hair and transparency!! :D We haven't seen any Live2D styled customizable models made for 3D yet, so we want to make a high quality model with lots of tech and features! -^u^- If you're interested, please let us know with VGen's 'get notified' feature!✨It would make all the effort Palette and I are putting into this all the more worthwhile!

Our customizable 3D vtuber model now has hair and transparency!! :D We haven't seen any Live2D styled customizable models made for 3D yet, so we want to make a high quality model with lots of tech and features! -^u^- If you're interested, please let us know with VGen's 'get notified' feature!✨It would make all the effort Palette and I are putting into this all the more worthwhile!

Pastell & Palette🤍 3D Customizable Model!

98,115 次观看 • 10 个月前

Bad at drawing different types of faces? With Clip Studio Paint's 3D head model, you can adjust parts like the face shape and eyes, or even change the angle. It's a great help when you want to draw a face shape you're unfamiliar with, or when you want to draw the same face repeatedly. For more information here ↓ #clipstudio

Bad at drawing different types of faces? With Clip Studio Paint's 3D head model, you can adjust parts like the face shape and eyes, or even change the angle. It's a great help when you want to draw a face shape you're unfamiliar with, or when you want to draw the same face repeatedly. For more information here ↓ #clipstudio

CLIP STUDIO PAINT

36,985 次观看 • 1 年前

✨ Made a new mini feature on Photo AI: [ Grab from 3d model ] So the problem is we're at that stage in time (typical for AI) where image-to-3d models are not good enough but are fun to play with, but we know they'll be good enough in 1-2 years With [ Make 3d model ] you already can turn any Photo AI pic into a 3d model but it still looks hyper clunky and deformed, but it works! One cool idea I had to make that more useful and made now: Let people make a 3d model then change the view of the it with the 3d viewer, then press [ o ] and it grabs a frame of the 3d That image you can then [ Remix ] (img2img), and it becomes a real photo again and that in turn you can then turn into a video again with [ Make video ] So that essentially gives you a fully freeform camera position control to take photos with One thing I need to fix is the background/skybox, I kinda need to take the original photo and remove the person and just get the background for the 3d model viewer, in this case it should be white, but it's a start!

✨ Made a new mini feature on Photo AI: [ Grab from 3d model ] So the problem is we're at that stage in time (typical for AI) where image-to-3d models are not good enough but are fun to play with, but we know they'll be good enough in 1-2 years With [ Make 3d model ] you already can turn any Photo AI pic into a 3d model but it still looks hyper clunky and deformed, but it works! One cool idea I had to make that more useful and made now: Let people make a 3d model then change the view of the it with the 3d viewer, then press [ o ] and it grabs a frame of the 3d That image you can then [ Remix ] (img2img), and it becomes a real photo again and that in turn you can then turn into a video again with [ Make video ] So that essentially gives you a fully freeform camera position control to take photos with One thing I need to fix is the background/skybox, I kinda need to take the original photo and remove the person and just get the background for the 3d model viewer, in this case it should be white, but it's a start!

@levelsio

119,210 次观看 • 1 年前

I know this is all over X but it’s fascinating. From #MeghanMarkle’s hand on the back of Harry’s neck forcing his head down to his obvious disinterest in kissing her-he’s far more interested in what’s in his bowl-there’s a lack of warmth between the 2 and a disconnect. You don’t see this interaction and think “this is a couple that is so in love”. It’s all performative. #WithLoveMeghan #HarryandMeghan

I know this is all over X but it’s fascinating. From #MeghanMarkle’s hand on the back of Harry’s neck forcing his head down to his obvious disinterest in kissing her-he’s far more interested in what’s in his bowl-there’s a lack of warmth between the 2 and a disconnect. You don’t see this interaction and think “this is a couple that is so in love”. It’s all performative. #WithLoveMeghan #HarryandMeghan

Princess CarParkle 👑

140,823 次观看 • 8 个月前

Apple just trained a 3D Gaussian head reconstruction model on 10,000+ subjects. Feed-forward. No test-time optimization. New identity in, reconstructed Gaussian head out. The UV-parameterized Gaussian representation decouples the number of Gaussians from the number and resolution of input images, making it practical to train with many high resolution views. And the heads are not just static either: text-conditioned identity generation, plus blendshape-driven latent animation across identities. We've been building in the 3D Gaussian Splatting space for a while. The gap between "research demo" and "works on real people at scale" is closing fast.

Apple just trained a 3D Gaussian head reconstruction model on 10,000+ subjects. Feed-forward. No test-time optimization. New identity in, reconstructed Gaussian head out. The UV-parameterized Gaussian representation decouples the number of Gaussians from the number and resolution of input images, making it practical to train with many high resolution views. And the heads are not just static either: text-conditioned identity generation, plus blendshape-driven latent animation across identities. We've been building in the 3D Gaussian Splatting space for a while. The gap between "research demo" and "works on real people at scale" is closing fast.

KIRI Engine - 3D Scanner App

12,181 次观看 • 2 个月前

Many people struggle with this “truth” … so many of you are struggling to draw the line, tell the truth … especially if it’s to someone you love or admire. This gentleman and many more in line at #veecon spoke on their inability to deliver a pushback .. I know it’s hard but candor when done kindly is a gift and it shouldn’t be confused with being angry or mean …

Many people struggle with this “truth” … so many of you are struggling to draw the line, tell the truth … especially if it’s to someone you love or admire. This gentleman and many more in line at #veecon spoke on their inability to deliver a pushback .. I know it’s hard but candor when done kindly is a gift and it shouldn’t be confused with being angry or mean …

Gary Vaynerchuk

34,685 次观看 • 2 年前

The low-quality dummy model of the Samsung Galaxy Z Fold 8 Wide has already appeared online. Its build quality is far from the real device, so it should only be used as a rough reference. From what I know, Samsung’s internal testers are actually very fond of this device. Maybe it’s because the form factor feels fresh and different.

The low-quality dummy model of the Samsung Galaxy Z Fold 8 Wide has already appeared online. Its build quality is far from the real device, so it should only be used as a rough reference. From what I know, Samsung’s internal testers are actually very fond of this device. Maybe it’s because the form factor feels fresh and different.

Ice Universe

105,330 次观看 • 2 个月前

I know I’ve joked about it a lot But seeing this on the big screen in an actual cinema full of people was pretty awesome :D Met so many cool ppl And have had sm more opportunities come from it all Super excited to share the next side quest, sooner than you’d think lmao

I know I’ve joked about it a lot But seeing this on the big screen in an actual cinema full of people was pretty awesome :D Met so many cool ppl And have had sm more opportunities come from it all Super excited to share the next side quest, sooner than you’d think lmao

Lukey

98,468 次观看 • 1 年前

50% cheaper Claude inference with just one line of code change! - Remove → model="claude-opus-4-8" - Add → model="ship-like/claude-opus-4-8" I verified the cost saving in my own terminal by invoking the same Anthropic model with the same prompt. The underlying engineering by Ship is actually interesting, and the patterns can be used in any production LLM stack. Essentially, a trained model is a frozen artifact. Every request performs the same forward-pass, whether it extracts a date or refactors a module, because the compute decision was made at training time, before the request existed. Ship makes that decision at inference time instead. After seeing a request, it searches over executions, involving single models, cascades, ensembles, or harnesses with tools, and serves the cheapest one that will match the reference model's quality. This is not a basic router, because picking a cheaper model per query doesn't ensure the cheaper model preserves the original's behavior, like output shape, tool-call patterns, and refusals. Ship measures this equivalence directly. Outputs stay distributionally indistinguishable from the reference model, not token-identical, since two calls to the same model already differ, but they are indistinguishable in capability and behavior. Of course, some requests execute cheaply and some cost Ship more than the customer pays, but the price per request is still a flat 50% off either way, so the execution-cost variance moves off the application's bill entirely. The video below depicts the cost savings and output in my real invocation, and I partnered with the team to put this together.

50% cheaper Claude inference with just one line of code change! - Remove → model="claude-opus-4-8" - Add → model="ship-like/claude-opus-4-8" I verified the cost saving in my own terminal by invoking the same Anthropic model with the same prompt. The underlying engineering by Ship is actually interesting, and the patterns can be used in any production LLM stack. Essentially, a trained model is a frozen artifact. Every request performs the same forward-pass, whether it extracts a date or refactors a module, because the compute decision was made at training time, before the request existed. Ship makes that decision at inference time instead. After seeing a request, it searches over executions, involving single models, cascades, ensembles, or harnesses with tools, and serves the cheapest one that will match the reference model's quality. This is not a basic router, because picking a cheaper model per query doesn't ensure the cheaper model preserves the original's behavior, like output shape, tool-call patterns, and refusals. Ship measures this equivalence directly. Outputs stay distributionally indistinguishable from the reference model, not token-identical, since two calls to the same model already differ, but they are indistinguishable in capability and behavior. Of course, some requests execute cheaply and some cost Ship more than the customer pays, but the price per request is still a flat 50% off either way, so the execution-cost variance moves off the application's bill entirely. The video below depicts the cost savings and output in my real invocation, and I partnered with the team to put this together.

Akshay 🚀

63,725 次观看 • 18 天前

Segment Any 3D Gaussians paper page: Interactive 3D segmentation in radiance fields is an appealing task since its importance in 3D scene understanding and manipulation. However, existing methods face challenges in either achieving fine-grained, multi-granularity segmentation or contending with substantial computational overhead, inhibiting real-time interaction. In this paper, we introduce Segment Any 3D GAussians (SAGA), a novel 3D interactive segmentation approach that seamlessly blends a 2D segmentation foundation model with 3D Gaussian Splatting (3DGS), a recent breakthrough of radiance fields. SAGA efficiently embeds multi-granularity 2D segmentation results generated by the segmentation foundation model into 3D Gaussian point features through well-designed contrastive training. Evaluation on existing benchmarks demonstrates that SAGA can achieve competitive performance with state-of-the-art methods. Moreover, SAGA achieves multi-granularity segmentation and accommodates various prompts, including points, scribbles, and 2D masks. Notably, SAGA can finish the 3D segmentation within milliseconds, achieving nearly 1000x acceleration compared to previous SOTA.

Segment Any 3D Gaussians paper page: Interactive 3D segmentation in radiance fields is an appealing task since its importance in 3D scene understanding and manipulation. However, existing methods face challenges in either achieving fine-grained, multi-granularity segmentation or contending with substantial computational overhead, inhibiting real-time interaction. In this paper, we introduce Segment Any 3D GAussians (SAGA), a novel 3D interactive segmentation approach that seamlessly blends a 2D segmentation foundation model with 3D Gaussian Splatting (3DGS), a recent breakthrough of radiance fields. SAGA efficiently embeds multi-granularity 2D segmentation results generated by the segmentation foundation model into 3D Gaussian point features through well-designed contrastive training. Evaluation on existing benchmarks demonstrates that SAGA can achieve competitive performance with state-of-the-art methods. Moreover, SAGA achieves multi-granularity segmentation and accommodates various prompts, including points, scribbles, and 2D masks. Notably, SAGA can finish the 3D segmentation within milliseconds, achieving nearly 1000x acceleration compared to previous SOTA.

AK

69,542 次观看 • 2 年前

American got a BBQ sandwich at a gas station and took it home to microwave it Shockingly the BBQ sauce isn’t a sauce at all, it’s transformed into a jelly like patty It’s a near solid, keeps its shape and flops around From what I can find it seems like many gas stations use commercial bbq sauces full of thickeners, they also use corn syrup and high high-fructose corn syrup combined can create this jelly Other people online think it’s 3D printed, less likely Our food is a science experiment. We have to Make America Healthy Again

American got a BBQ sandwich at a gas station and took it home to microwave it Shockingly the BBQ sauce isn’t a sauce at all, it’s transformed into a jelly like patty It’s a near solid, keeps its shape and flops around From what I can find it seems like many gas stations use commercial bbq sauces full of thickeners, they also use corn syrup and high high-fructose corn syrup combined can create this jelly Other people online think it’s 3D printed, less likely Our food is a science experiment. We have to Make America Healthy Again

Wall Street Apes

91,603 次观看 • 3 个月前

✨ Every week a new AI model comes out and it suddenly makes my half broken features work a lot better Yesterday Seedream-4-Edit came out and it made my [ Hold product ] feature on Photo AI a lot better You can now go from: 🎁 Product photo -> 👱‍♀️ Talking video with your AI model while holding your product. In just a few minutes! Here's a photo I took from the weekly farm box we get in our kitchen, I set it as the product and then with Photo AI made it into a talking video where my trained AI model presents it It's not perfect, as the objects inside the farm box still move around a bit, but pretty close. If the product is more uniform (like lip gloss, a product box or a book) it does a pretty good job at keeping it exactly the same This "consistency" as they call it is quite important for actual real world use. Product sellers don't want to have an image or video of an AI model if the product doesn't look exactly the same as what they sell With that, I'm getting pretty close now and every week with every new model that comes out, a bit closer And it's interesting cause now I'm finally moving from B2C a bit more to B2B where businesses can use Photo AI more, designers and stores already use it for trying on clothes etc. but now they can generate content for real products! 😊 LIVE now on Photo AI

✨ Every week a new AI model comes out and it suddenly makes my half broken features work a lot better Yesterday Seedream-4-Edit came out and it made my [ Hold product ] feature on Photo AI a lot better You can now go from: 🎁 Product photo -> 👱‍♀️ Talking video with your AI model while holding your product. In just a few minutes! Here's a photo I took from the weekly farm box we get in our kitchen, I set it as the product and then with Photo AI made it into a talking video where my trained AI model presents it It's not perfect, as the objects inside the farm box still move around a bit, but pretty close. If the product is more uniform (like lip gloss, a product box or a book) it does a pretty good job at keeping it exactly the same This "consistency" as they call it is quite important for actual real world use. Product sellers don't want to have an image or video of an AI model if the product doesn't look exactly the same as what they sell With that, I'm getting pretty close now and every week with every new model that comes out, a bit closer And it's interesting cause now I'm finally moving from B2C a bit more to B2B where businesses can use Photo AI more, designers and stores already use it for trying on clothes etc. but now they can generate content for real products! 😊 LIVE now on Photo AI

@levelsio

361,558 次观看 • 11 个月前

not sure why nobody is talking about this but Google Omni is insane at video editing Original Video (left) vs Omni Edited Video (right) everyone is comparing it to Seedance and missing the point completely. Seedance is for generating videos from scratch. Google Omni is for editing videos that already exist. which are two completely different use cases this is like when Nano Banana 1 first came out and nobody realized how big it was going to be. this is the first AI that can actually properly edit videos.. i've generated a few hundred videos with this model and it can do literally any type of edit you can think of. changing voices, swapping characters, removing watermarks, adding captions, transitions, pop ups, whatever. if you can describe the edit you want it can do it this completely crushes every other model on the market when it comes to video editing. nothing else even comes close right now and this is just the flash model. imagine what the pro version is going to be able to do when it drops in a couple months this should have way more hype than it's getting..

not sure why nobody is talking about this but Google Omni is insane at video editing Original Video (left) vs Omni Edited Video (right) everyone is comparing it to Seedance and missing the point completely. Seedance is for generating videos from scratch. Google Omni is for editing videos that already exist. which are two completely different use cases this is like when Nano Banana 1 first came out and nobody realized how big it was going to be. this is the first AI that can actually properly edit videos.. i've generated a few hundred videos with this model and it can do literally any type of edit you can think of. changing voices, swapping characters, removing watermarks, adding captions, transitions, pop ups, whatever. if you can describe the edit you want it can do it this completely crushes every other model on the market when it comes to video editing. nothing else even comes close right now and this is just the flash model. imagine what the pro version is going to be able to do when it drops in a couple months this should have way more hype than it's getting..

Miko

29,783 次观看 • 2 个月前

If you typically stream with a 3D model, I highly recommend that you pose your model when you’re not doing full body mocap instead of letting it stay in the stiff generic pose! It makes a huge difference in how your energy is conveyed (ᗒ⩊ᗕ)⸝ި ʕᦏ⌎ I’m only saying this because I noticed this a lot with my own 3D kids, but here’s a comparison showcase: ← default pose & default pendulum physics in Warudo → custom poses & slightly adjusted pendulum physics If you find it hard to pose within Warudo by using bone offsets or none of the existing poses in Warudo vibe with you, you can make your own like me! I made my custom poses in “VRM Posing Desktop” (this is the BEST vrm posing app I’ve ever used in the past 3 years) and exported them as Unity anims then dropped them into Warudo’s animation folder! You can then make a simple blueprint in Warudo to toggle between poses and make your model look more alive! It should fit especially well for just chatting streams 🙂‍↕️✨

If you typically stream with a 3D model, I highly recommend that you pose your model when you’re not doing full body mocap instead of letting it stay in the stiff generic pose! It makes a huge difference in how your energy is conveyed (ᗒ⩊ᗕ)⸝ި ʕᦏ⌎ I’m only saying this because I noticed this a lot with my own 3D kids, but here’s a comparison showcase: ← default pose & default pendulum physics in Warudo → custom poses & slightly adjusted pendulum physics If you find it hard to pose within Warudo by using bone offsets or none of the existing poses in Warudo vibe with you, you can make your own like me! I made my custom poses in “VRM Posing Desktop” (this is the BEST vrm posing app I’ve ever used in the past 3 years) and exported them as Unity anims then dropped them into Warudo’s animation folder! You can then make a simple blueprint in Warudo to toggle between poses and make your model look more alive! It should fit especially well for just chatting streams 🙂‍↕️✨

𝗞𝗔𝗥𝗜𝗛𝗔 🌘🍀 3D Artist ☻

15,700 次观看 • 5 个月前

The S3XY Dash from Enhance - S3XY Buttons is one of the coolest Tesla accessories I've seen yet. I know many people want Android Auto/Apple Car Play, and this gives the OPTION to use those with your Tesla, along with tons of other info. They have 20% off for preorder, and you can get an additional 10% off with my link: I love using Waze on road trips, and this looks like the way to do it going forward. On top of having a customizable sized android auto experience, the Dash can display charging, vehicle, motor, battery and tons of other info. Enhance was also smart enough to not force it directly in front of the air vents. You can place it where you'd like with their included kit. All the widgets are also customizable, so you can add any you like and resize them. You can also make multiple pages to quickly flip through different info. There's SO much you can do with this *if you want it*. Great job on this one.

The S3XY Dash from Enhance - S3XY Buttons is one of the coolest Tesla accessories I've seen yet. I know many people want Android Auto/Apple Car Play, and this gives the OPTION to use those with your Tesla, along with tons of other info. They have 20% off for preorder, and you can get an additional 10% off with my link: I love using Waze on road trips, and this looks like the way to do it going forward. On top of having a customizable sized android auto experience, the Dash can display charging, vehicle, motor, battery and tons of other info. Enhance was also smart enough to not force it directly in front of the air vents. You can place it where you'd like with their included kit. All the widgets are also customizable, so you can add any you like and resize them. You can also make multiple pages to quickly flip through different info. There's SO much you can do with this if you want it. Great job on this one.

Dirty Tesla

59,866 次观看 • 8 个月前

If you think this is just another silly demo made with AI, read this post. You might change your mind, because this demo is about MATH. What you see on the screen is not a render from Blender (obviously, it’s not that good). It’s a three.js app built with Toolcraft. Available on the web and rendered in real time(link in the comments). But Blender still has a lot to do with it. Blender has Geometry Nodes - a powerful node-based system for creating and manipulating procedural geometry. In other words, it’s math. And math is a universal language. And who do you think is pretty good at math? >>> AI. Now you can download or buy Blender files from marketplaces, and when they contain Geometry Nodes for procedural animations, objects, surfaces, or effects, you can transfer that logic to the web. Make it real-time, make it interactive. Materials are a separate story, of course. They can still suck unless you use the right tricks: PBR, HDRIs, material blending, displacement, and faked surface relief. So why is Blender important here? Blender is open source, and many tools around it are open source too. An AI trained on their code. That means it can translate the math from one environment to another quite accurately. If you’ve been struggling to reproduce some idea with AI that you had in your head or seen in some references, and it has something to do with Geometry Nodes, and you can find that idea or a close one in the Blender ecosystem - it means you can transfer it to the web. Thank me later.

If you think this is just another silly demo made with AI, read this post. You might change your mind, because this demo is about MATH. What you see on the screen is not a render from Blender (obviously, it’s not that good). It’s a three.js app built with Toolcraft. Available on the web and rendered in real time(link in the comments). But Blender still has a lot to do with it. Blender has Geometry Nodes - a powerful node-based system for creating and manipulating procedural geometry. In other words, it’s math. And math is a universal language. And who do you think is pretty good at math? >>> AI. Now you can download or buy Blender files from marketplaces, and when they contain Geometry Nodes for procedural animations, objects, surfaces, or effects, you can transfer that logic to the web. Make it real-time, make it interactive. Materials are a separate story, of course. They can still suck unless you use the right tricks: PBR, HDRIs, material blending, displacement, and faked surface relief. So why is Blender important here? Blender is open source, and many tools around it are open source too. An AI trained on their code. That means it can translate the math from one environment to another quite accurately. If you’ve been struggling to reproduce some idea with AI that you had in your head or seen in some references, and it has something to do with Geometry Nodes, and you can find that idea or a close one in the Blender ecosystem - it means you can transfer it to the web. Thank me later.

Alex Barashkov

28,090 次观看 • 23 天前

Most people still underestimate memecoins. Where else can a newly launched project can go from a $50K Mcap to $500M in few days or weeks? We've seen few recent memecoins like $PUNCH, $FARTCOIN, $GIGA, $TROLL, and many more go from tiny mcap to tens or even hundreds of millions in record time. You won't find that kind of upside in stocks, real estate, or even most large-cap crypto assets. Yes, the risks are high, but ignoring the opportunity completely can be just as costly. The biggest gains usually go to those who spot the trend before everyone else. #Memecoins #1000x #Pumpfun

Most people still underestimate memecoins. Where else can a newly launched project can go from a $50K Mcap to $500M in few days or weeks? We've seen few recent memecoins like $PUNCH, $FARTCOIN, $GIGA, $TROLL, and many more go from tiny mcap to tens or even hundreds of millions in record time. You won't find that kind of upside in stocks, real estate, or even most large-cap crypto assets. Yes, the risks are high, but ignoring the opportunity completely can be just as costly. The biggest gains usually go to those who spot the trend before everyone else. #Memecoins #1000x #Pumpfun

VikingLord.eth

58,561 次观看 • 1 个月前

Transformer by hand ✍️ ~ 6 steps walkthrough below Open the hood of a transformer and the parts list is overwhelming: embeddings, positional encoding, attention weighting, self-attention, cross-attention, multi-head attention, layer norm, skip connections, softmax, linear, Nx, shifted right, query, key, value, masking. Which of those actually make the car run? Two of them. Attention weighting and the feed-forward network. Everything else is an enhancement to make it run faster and longer, which is how we got from a car to a truck, and to the word "large" in large language model. So I drew and calculated those two parts entirely by hand. Goal: push five features through one transformer block, filling in every cell yourself. 1. Given Five positions of input features, arriving from the previous block. 2. Attention matrix Let us feed all five features to a query-key module (QK) and read back an attention weight matrix, A. The details of that module are a post of their own. 3. Attention weighting We multiply the input features by A to get the attention weighted features, Z. Still five positions. The effect is to combine features *across positions*, horizontally: X1 becomes X1 + X2, X2 becomes X2 + X3, and so on. 4. First layer Let us feed all five weighted features into the first layer of the FFN. Multiply by the weights and biases. This time the combining happens *across feature dimensions*, vertically, and each feature grows from 3 numbers to 4. Note that every position goes through the same weight matrix. That is what "position-wise" means. 5. ReLU We cross out the negatives. They become zeros. 6. Second layer Let us bring it back down: 4 dimensions to 3. The output feeds the next block, which has a completely separate set of parameters, and the whole thing runs again. You have just calculated a transformer block by hand. ✍️ The takeaway: the two parts are doing two different jobs, and neither one alone is enough. Attention mixes *across positions*, so a feature can see its neighbours. The FFN mixes *across feature dimensions*, so each position can think about itself. Horizontal, then vertical. Then that pattern repeats N times, each block with its own separate set of weights. That is the Nx from the list up top, and that is what makes the transformer run. 💾 Save this post! #AIbyHand #Transformers #DeepLearning

Transformer by hand ✍️ ~ 6 steps walkthrough below Open the hood of a transformer and the parts list is overwhelming: embeddings, positional encoding, attention weighting, self-attention, cross-attention, multi-head attention, layer norm, skip connections, softmax, linear, Nx, shifted right, query, key, value, masking. Which of those actually make the car run? Two of them. Attention weighting and the feed-forward network. Everything else is an enhancement to make it run faster and longer, which is how we got from a car to a truck, and to the word "large" in large language model. So I drew and calculated those two parts entirely by hand. Goal: push five features through one transformer block, filling in every cell yourself. 1. Given Five positions of input features, arriving from the previous block. 2. Attention matrix Let us feed all five features to a query-key module (QK) and read back an attention weight matrix, A. The details of that module are a post of their own. 3. Attention weighting We multiply the input features by A to get the attention weighted features, Z. Still five positions. The effect is to combine features across positions, horizontally: X1 becomes X1 + X2, X2 becomes X2 + X3, and so on. 4. First layer Let us feed all five weighted features into the first layer of the FFN. Multiply by the weights and biases. This time the combining happens across feature dimensions, vertically, and each feature grows from 3 numbers to 4. Note that every position goes through the same weight matrix. That is what "position-wise" means. 5. ReLU We cross out the negatives. They become zeros. 6. Second layer Let us bring it back down: 4 dimensions to 3. The output feeds the next block, which has a completely separate set of parameters, and the whole thing runs again. You have just calculated a transformer block by hand. ✍️ The takeaway: the two parts are doing two different jobs, and neither one alone is enough. Attention mixes across positions, so a feature can see its neighbours. The FFN mixes across feature dimensions, so each position can think about itself. Horizontal, then vertical. Then that pattern repeats N times, each block with its own separate set of weights. That is the Nx from the list up top, and that is what makes the transformer run. 💾 Save this post! #AIbyHand #Transformers #DeepLearning

Tom Yeh

25,944 次观看 • 23 天前

first channel i spotted doing this this video generated +260k views in ~30days his strategy consists of animating only the first 1.5 min of the video: > ppl preview his vid while scrolling the YT feed > they see it’s animated == they click > after 2m they’re already engaged > animations disappear > but users keep watching until the end this method seems so obvious to me, don’t know why more people aren’t doing it you save around 3-4x of the costs of animating all the clips while increasing 1.5-2x times the chances of getting better results more engagement more views more revenue from AdSense on Calliopelabs you can do that with full control while still saving several hours of manual generations i recommend using the new Low model, it’s VERY cheap ($0.03/sec) (720p + sfx!) and it’ll get you his same or even better results it’s s bad that i say this but i cooked with this one 🔥

first channel i spotted doing this this video generated +260k views in ~30days his strategy consists of animating only the first 1.5 min of the video: > ppl preview his vid while scrolling the YT feed > they see it’s animated == they click > after 2m they’re already engaged > animations disappear > but users keep watching until the end this method seems so obvious to me, don’t know why more people aren’t doing it you save around 3-4x of the costs of animating all the clips while increasing 1.5-2x times the chances of getting better results more engagement more views more revenue from AdSense on Calliopelabs you can do that with full control while still saving several hours of manual generations i recommend using the new Low model, it’s VERY cheap ($0.03/sec) (720p + sfx!) and it’ll get you his same or even better results it’s s bad that i say this but i cooked with this one 🔥

Sergio Gil

33,576 次观看 • 4 个月前

🌀 #Live3D #Live2D #Live2DWIP 🌀 Today, I’d like to share two of the three methods I use to create joint twisting in my Live2D-based 2D Pseudo-3D concept. On the right is the Glue Joint： The basic idea is to use Glue to connect two pseudo-3D ArtMeshes. By taking advantage of Glue’s weight settings, it can create an effect similar to a 3D joint. Once the weights are set up, it can move very freely with a relatively low amount of work. However, it also shares the same weakness as 3D joints: when rotated at large angles, it can produce unpleasant twisting and distortion. On the left is the Hand-Bent Joint： The basic idea is to add bending Keyforms directly to a Warp Deformer, then manually sculpt the ideal deformation. This method can create the most detailed and refined joints, depending entirely on how the rigger designs them. The downside is that it requires much more work. Every additional degree of freedom requires a significant amount of manual adjustment. These two joint methods have almost opposite characteristics, so in actual Live3D work, I choose the most suitable method depending on the situation. For example, in this hand model, I used Hand-Bent Joints for the arm twisting and fingers, while the wrist uses a Glue Joint.✨

🌀 #Live3D #Live2D #Live2DWIP 🌀 Today, I’d like to share two of the three methods I use to create joint twisting in my Live2D-based 2D Pseudo-3D concept. On the right is the Glue Joint： The basic idea is to use Glue to connect two pseudo-3D ArtMeshes. By taking advantage of Glue’s weight settings, it can create an effect similar to a 3D joint. Once the weights are set up, it can move very freely with a relatively low amount of work. However, it also shares the same weakness as 3D joints: when rotated at large angles, it can produce unpleasant twisting and distortion. On the left is the Hand-Bent Joint： The basic idea is to add bending Keyforms directly to a Warp Deformer, then manually sculpt the ideal deformation. This method can create the most detailed and refined joints, depending entirely on how the rigger designs them. The downside is that it requires much more work. Every additional degree of freedom requires a significant amount of manual adjustment. These two joint methods have almost opposite characteristics, so in actual Live3D work, I choose the most suitable method depending on the situation. For example, in this hand model, I used Hand-Bent Joints for the arm twisting and fingers, while the wrist uses a Glue Joint.✨

📐Hephaestus📏Live2D匠人魂

18,304 次观看 • 1 个月前