正在加载视频...

视频加载失败

✨#Warudo New Feature: Camera Background Input 2D 背景をカスタマイズしよう! 🎇Now you can freely add 2D background to the Warudo interface! This includes custom Images, Videos, Spout sources, and NDI sources. This setting allows the background to remain in a fixed position, unaffected by the control of any other assets. Simply...

12,692 次观看 • 4 天前 •via X (Twitter)

9 条评论

Manta マンタ 🔌🫧 3DVTuber 的头像
Manta マンタ 🔌🫧 3DVTuber3 天前

Big things happening

Kaori Lupine 💢🔥 || VTuber 的头像
Kaori Lupine 💢🔥 || VTuber4 天前

Good stuff 👏

Yena Youngblood ⛓️‍💥 的头像
Yena Youngblood ⛓️‍💥4 天前

warudo my goat

Kimii Kimura|❔🎀 的头像
Kimii Kimura|❔🎀4 天前

YESSSSSS

Tipanya 🐻🐱 的头像
Tipanya 🐻🐱4 天前

YESSSS THANK YOU SO MUCH 😭😭😭

rial miruku 🪷💚【vtuber】 的头像
rial miruku 🪷💚【vtuber】4 天前

oh my gosh yes finally! this is great 🎉🎉🎉

♆ tricky koopa ♆ 的头像
♆ tricky koopa ♆4 天前

AAAAAA THIS IS SO NEXT LEVEL 😭😭😭😭 @kayaterushi I can walk around in your backgrounds now!!!

AltariaBuildCrew - Vtubers 的头像
AltariaBuildCrew - Vtubers4 天前

🎉

The Ugly Bastard Vtuber DOOMBOT 的头像
The Ugly Bastard Vtuber DOOMBOT4 天前

Okay, did they fix the Controller/Mouse and Keyboard Issues?

相关视频

CSS Trick! ⚡️ You can use scroll-driven animation with background-attachment to create a dynamic glowing card scroller without JS 🔥 section { animation:vibe; animation-timeline:--list; } @​keyframes vibe { to{--hue:320;}} .glow {background: hsl(var(--hue) 80% 50%);} Here's how! 🤙 You can use the background-attachment trick used in other glow card demos 😎 article { background-attachment: fixed; } The difference here is that you aren't going to update the fixed background position with your pointer this time. It can remain fixed. The magic part is that as you scroll, the background will leave the card that's leaving and enter the card that's entering ✨ For the extra background glow, you can use a fixed pseudo element on the list container itself 💪 Once that's in place, you're only task is to change the color of the background as you scroll 🤔 Create a custom property declaration for the --hue @​property --base { inherits: true; syntax: ' '; initial-value: 0; } Then create an animation that updates this value @​keyframes accent { to { --hue: 320; }} The last piece is hooking it up to scroll and there is a little trick in here 👀 First, you need an inline scroll-timeline on the list ul { scroll-timeline: --list inline; } Then you can use timeline-scope to hoist that scroll-timeline up so a parent can use it. You then animate the custom property on this element and let the value cascade down to the places that need it 🔥 section { timeline-scope: --list; animation: accent both linear; animation-timeline: --list; } For example, the glow uses the --hue this way [data-glow] { background-image: radial-gradient( 150px 150px at 50% 50%, hsl(var(--hue) 100% 70% / 0.25), transparent ); } Lastly, scroll-snap is optional of course but plays nice with the scroll-driven animation demos ✨ The key for that is ul { scroll-snap-type: x mandatory; } li { scroll-snap-align: center; } That's it! Pretty fun trick to play with! 🤓 Any questions, let me know! Should we add it to the video walkthrough list? CodePen.IO link below! 👇

jhey ʕ•ᴥ•ʔ

116,625 次观看 • 2 年前

📢📢 𝐀𝐯𝐚𝐭𝟑𝐫 📢📢 Avat3r creates high-quality 3D head avatars from just a few input images in a single forward pass with a new dynamic 3DGS reconstruction model. Video: Project: Our core idea is to make Gaussian Reconstruction Models animatable. We find that a simple cross-attention to an expression code sequence is already sufficient to model complex facial expressions. We then incorporate position maps from DUSt3R and feature maps from Sapiens to facilitate the prediction task. While DUSt3R's position maps act as a pixel-aligned initialization for the Gaussians' positions, the Sapiens feature maps help the cross-view transformer to match corresponding image tokens in the 4 input images. One major challenge in creating a 3D head avatar from smartphone images comes from inconsistent facial expressions when the subject could not remain perfectly static during the capture. We eliminate this static requirement by simply showing our model input images with different facial expressions during training. This technique makes our model robust to inconsistent input images later on. Finally, we show that despite the model has been trained with 4 input images, one can even create a 3D head avatar when only a single image is available. To achieve this, we employ a pre-trained 3D GAN to lift the single image to 3D and then render the 4 input images for our model. This allows us to create 3D head avatars from single images and even highly out-of-distribution examples like AI generated faces, paintings or statues. Great work by Tobias Kirschstein from his internship at Meta with Javier Romero, Artem Sevastopolsky, and Shunsuke Saito

Matthias Niessner

74,818 次观看 • 1 年前

Got flash banged by OpenCode recently? Yeah, that was me. Here's what happened: OpenCode/OpenTUI detects terminal theme mode (light or dark) by querying OSC 10/11 for the terminal’s foreground/background colors and deriving the mode from those values. OpenTUI also enables DEC Private Mode 2031 to receive theme-change notifications. When that mode is enabled, terminals can send CSI 997 when the system theme changes. OpenTUI treated the 997 payload as authoritative and let it override the OSC 10/11-derived result. That turned out to be wrong. In some terminals, the 997 payload is simply unreliable: it can report light while the queried terminal colors clearly indicate a dark theme. That’s what caused the flash bangs. The fix seemed straightforward: stop trusting the 997 payload as the actual theme mode and use it only as a trigger to re-query OSC 10/11, then derive the mode from the returned foreground/background colors. That works great. Except OpenTUI had also started using OSC 11 to set the terminal’s default background color, so the terminal gutter around the drawn TUI matched the app background instead of the terminal’s own default background. This was introduced recently in OpenTUI. That created a new problem: if you use OSC 11 to override the terminal background, then later query OSC 11, you are just reading back your own override instead of the terminal’s real theme background. No problem. Reset the terminal background with OSC 111, query OSC 10/11, derive the mode, then restore the renderer background again. Oh wait. Turns out Ghostty has a bug. Once this background reset/override path is used, later OSC 11 background reporting can get stuck and stop tracking system theme changes correctly. Even though theme-change notifications still fire and OSC 10 foreground updates continue. This would cause further issues where the OSC 10/11 derivation would be wrong again. So for now: no using OSC 11 to set the terminal background color just to fill the gutter. TL;DR: OpenTUI must treat CSI 997 only as a hint that the terminal theme may have changed and rely exclusively on OSC 10/11 color queries for actual theme mode detection, while avoiding terminal background mutation via OSC 11/111 because that breaks correctness in some terminals like Ghostty.

kmdr

30,311 次观看 • 5 个月前