Nicholas

Blending soulful art with neural network sorcery 🔥
Teaching machines to feel while falling in love with music, film, and the stories cultures tell #ai 🐾

[Stable Diffusion Tutorials @SD_Tutorial]

Krea 2 🤩 styles (wildcards txt)

G-Drive/pastebin links (files)
in comment section👇😃

[Nicholas @Celtya]

@twinewss Не понимаю, при чем тут розовые очки, дискриминация, о чем народ пишет. Дети просто искренно повтор��ют выражения, а взрослы с учетом жизненного опыта уже различают как черное и белое, и возможно понимают где гримаса осознанная, а где результат другого характера, нюансы есть конеч

[Nicholas @Celtya]

@ltx_io @ComfyUI I rather had another problem, a long loading time after each new prompt... Of course it’s more convenient in ComfyUi, although I used something else like Wan2Gp

[Voxon @voxonphotonics]

Human performance and volumetric display are a powerful combination.

Using FaceAnything, we converted iconic movie performances into volumetric content, preserving the subtle expressions and emotion that make great acting so compelling.

[email protected]

#VolumetricDisplay

[el.cine @EHuanglu]

omg.. AI just fixed b&w silent films

this AI video composer can analyse the story and generate SFX and music for entire film.. and chat with agent to add and edit

check out workflow:

[Nicholas @Celtya]

@teren0894 nope, sry 😰

[VoxelKei @VoxelKei]

3D Gaussian Splattingの新しい見せ方を作りました。その名も Inversphere.
2Dのリトルプラネットと違い、空間ごと球面反転しています。ガラス玉に入れる従来の見せ方は球で「切り取る」のに対して、これは球の中へ「折り込む」ので360度の景色が無限遠まで立体で入っています。
#3DGS

[Nicholas @Celtya]

@Stylez_Morales awesome! 👍

[Nicholas @Celtya]

@EHuanglu Skynet..4, 3, 2, 1... ))

[Decart @DecartAI]

Meet Lucy 2.5, our most advanced Live AI model yet.

Lucy edits videos in realtime, now with more capabilities and greater control.

See how it's being used across streaming, e-commerce, advertising, and more 🧵

[Nicholas @Celtya]

@pressmanc I hope everything will be fine!

[Nicholas @Celtya]

@icreatelife

[Nicholas @Celtya]

@AleRVG 😁👏

[el.cine @EHuanglu]

wow.. AI now can do color restoration accurately and 4k upscale for old films..

[Justine Moore @venturetwins]

AI video is remarkably good at imagining historical events.

This is a one-shot video from Seedance 2 of Pompeii when Mt. Vesuvius erupted.

Prompt is below, you can ask an LLM to rewrite it for any other event 👇

This is handheld documentary footage recorded on an early-2000s consumer DV camcorder from the streets of Pompeii at the moment Mount Vesuvius begins erupting. The footage feels like real, imperfect home video of ordinary people witnessing the beginning of a historic disaster.

The recording shows a busy street in Pompeii during the daytime, with merchants, families, children, laborers, and townspeople moving through the city. In the distance, Mount Vesuvius is visible above the rooftops. At first, people are going about normal daily life, but then a strange plume of smoke and ash begins rising from the volcano. The crowd gradually notices it. Some people stop and stare, some point toward the mountain, and others begin shouting to one another in confusion.

The camera moves through the street like someone in the crowd trying to capture what is happening. It shows frightened townspeople looking up at the sky, parents pulling children closer, merchants abandoning their stalls, and groups of people beginning to run as ash starts falling. The footage captures the shift from curiosity to fear as the eruption becomes impossible to ignore. There are natural cuts between wider views of the street, the mountain in the distance, and closer views of people reacting in panic.

The handheld camera shows natural shake, drifting framing, sudden reactive movement, autofocus mistakes as the person filming swings between the crowd and the volcano, slight exposure problems from bright daylight and ash in the air, and the imperfect look of old DV footage. The movement should feel urgent and unplanned, like someone trying to document the disaster while also staying safe.

Natural sound only: the noise of the street, footsteps on stone, people murmuring and then shouting, children crying, distant rumbling from the volcano, debris falling, and the panic of the crowd. No cinematic music added.

The result must feel like authentic, raw footage of ordinary people in Pompeii experiencing the beginning of the eruption of Mount Vesuvius, captured on an old DV camcorder.”

[el.cine @EHuanglu]

wow.. AI agents can do 4K VFX for you now

connect Kling MCP to OpenClaw or Hermes, the agent controls your computer to key green screen, replace bg, swap actors, add VFX by itself

here’s how to install and use it:

[A.I.Warper @AIWarper]

Testing Seedance vs Google Omni for VFX style stuff

Seedance is vastly superior in every test I have done.

Prompt:

"Replace the background with an enormous sea creature pursuing the boat. Have it end with the creature catching up and just about to swallow the boat. Match the exact verbatim camera angle and position of the boat from the reference video."

[el.cine @EHuanglu]

what if you could see life 900 years ago in 4k?

i asked AI, and it wrote the story, the script, every prompt, then turned it into a 4K film in just a few hrs using Seedance 2 on BytePlus

prompts below:

[Linoy Tsaban @linoy_tsaban]

Meet the LTX 2.3 Storyboard playground 🎬

LTX folks made a great point about text-to-video models use case, so I turned their practical guide into a ZeroGPU space for fast, cheap iterations before full video gen
> iterate on your storyboard quickly
> easy to customize (duplicate the space and build on top for free)
> it's really fun

try it: https://t.co/Ldjikdusy2

[Nicholas @Celtya]

@ArcVyck The gloss and reflections are terrible; it’s impossible to watch properly.