gerhard koenderink

creative studio | pixels, pulp + canvas

[aicreataro @aicreataro]

「Dot battle」
バトル素材で違った表現を模索。
Opus 5.5でBlenderとPythonとAEを操作。

AIの映像を拡張できそうな予感。使い所は不明。
でもドットを打つことが大事ってジョブズも言ってた。

■素材
・バトル映像と音：Seedance 2.0

■Blenderとコードでやらせたこと
・人物の輪郭と奥行きを解析して、映像をドットに置き換える
・半拍ごとに、ドットの大きさ・色・背景を切り替える
・打撃に合わせて、ドットが潰れる、カメラへ飛び散る、時間が止まってカメラが回り込む

■わかったこと
・ドットが大きすぎると、何の映像か伝わらない。小さすぎるとかっこ良くない。
・切り替えを1拍ごとから半拍ごとにして、打撃で粒を潰したら、一気に激しくなった
・ずっと速いだけだと単調。止める場面を挟むと、速い所が効く
・打撃のたびにドットを爆発させたら、人物が読めなくなった。爆発は弱く短く

[Kōda @aimikoda]

Seedance 2.0 Opera with Phonetics (IPA)

/haɪ aɪ sɪŋ ˈɑpərə/

You can find the prompt below 👇

[Parul Gautam @Parul_Gautam7]

https://t.co/JlD2SxErLM

[TechHalla @techhalla]

Local. Open weights. Native 4K.

LTX-2 is now a 100% Open Source AI video model, and I tested it on my rig!

Installation, VRAM usage, and the prompts I used for this video, below 👇

[AI Search @aisearchio]

Open-source SUNO is here!

Introducing HeartMula: free & offline AI music generator

Full tutorial: https://t.co/Z1FBCpmXUb

[Wildminder @wildmindai]

Move over, Suno. There’s a new music model, definitely need to try it - HeartMuLa.
- open-source;
- LLM-based song gen, multi-modal inputs, section-specific style prompts;
- tops Suno v5 & Udio v1.5 on lyric clarity.
https://t.co/JQE8jAJ2V1

[Purz.ai @PurzBeats]

LTX-2 t2v & i2v Workflows
https://t.co/zClpBRf6z3

AnimateDiff LoRA + Workflow by @nebsh83
https://t.co/9WkBViecGk

[Ege @egeberkina]

Turn one hero image into a full multi-angle carousel without writing a single prompt.

Front, 3/4, top down, low angle and all from a single image.

Qwen Edit Angles in Flora is kind of crazy. You can just rotate the camera and keep the subject exactly the same.

Workflow 👇

[willie @ReflctWillie]

I improved my workflow for contact sheet prompting using Nano Banana Pro. All the prompts, workflow files, and tips below.

These are focused on 'fashion style' shoots, but can be adapted to any I2V workflow.

[Henry Daubrez 🌸💀 @henrydaubrez]

THE LIGHT WITHIN

A concept trailer powered by Nano Banana Pro 🍌

I haven’t been impressed by new AI tools lately but this one swept me off my feet.

The kicker?
I used Google Flow and a few wild tricks to get perfect consistency.

Here's a step-by-step full breakdown 🧵

#1/15

[Techartist @techartist_]

I pasted the @threejs code from the previous 3D scene in Grok 4 from @xai and prompted it to create a variation with new shapes. The scene leverages custom GLSL shaders for particle behavior and click morphs between shapes. Open source code and workflow in the thread below. 🧵👇

[Emily @IamEmily2050]

Google Veo3 fast, on Google Gemini App 😀

{
"concept": "Inside a Korean supermarket at night, a 23-year-old Korean woman and a pixelated 16-bit Yeti perform a trending TikTok dance in the middle of the aisle. Shoppers stop and watch in awe, amusement, and disbelief as the real and digital worlds collide.",
"shot": {
"composition": "medium-wide frontal shot framed down the aisle, centered on the dancers, depth created by store shelves and scattered onlookers",
"camera_motion": "slow lateral dolly or static with slight handheld feel to mimic organic bystander filming",
"frame_rate": "30fps",
"film_grain": "minimal, clean digital look with soft texture"
},
"subject": {
"Character_1": {
"description": "23-year-old Korean woman with black ponytail, wearing a bright cropped hoodie and black leggings with white sneakers, energetic and confident"
},
"Character_2": {
"description": "16-bit pixelated Yeti, chunky and expressive, white fur blocks, cyan glowing eyes, perfectly mirroring her choreography with a cartoon twist"
}
},
"scene": {
"location": "well-lit Korean supermarket, night time, convenience-style layout with narrow aisles and fluorescent lighting",
"time_of_day": "night",
"environment": "brightly stocked shelves with snacks, noodles, and drinks. Neon sale signs, open refrigerators buzzing softly, and a few scattered baskets and carts"
},
"visual_details": {
"action": "The woman and pixel Yeti perform a precise, upbeat TikTok routine in sync. Around them, shoppers pause to watch—some filming on phones, others reacting with surprise or laughter. One child tries to copy the moves. The Yeti adds a final pixelated flourish at the end, tossing pixel confetti",
"props": "shopping carts, stacked ramen boxes, sale banners, refrigerated drinks glowing with light, smartphones raised by onlookers"
},
"cinematography": {
"lighting": "fluorescent top-down supermarket lights, flat and clean, with some natural bounce from packaging and product labels",
"tone": "fun, energetic, surreal slice-of-life with digital twist"
},
"audio": {
"music": "upbeat viral TikTok track with crisp beat drops",
"sfx": "grocery cart wheels, phone camera shutters, subtle refrigerator hum, faint crowd murmurs and laughter, pixel sparkles with each Yeti move"
},
"color_palette": "bright commercial tones—white light, colorful product packaging, primary clothing colors, soft neon accents from packaging and coolers"
}

[DANN© @DannPetty]

Created with Midjourney + ChatGPT + Figma in just under one hour.

Quick tutorial how to make it yourself👇

[Martin Nebelong @MartinNebelong]

Character consistency in VEO3 I2V? The trick? The same start-frame for all clips and: "Instantly jump/cut on frame 1 to [describe the new scene loosely]. [Add a more detailed description of the new scene].

The song is "I'll Understand" by Roza, from https://t.co/XYXC9ovqKE.

[el.cine @EHuanglu]

wow.. this is crazy

Higgsfield AI just dropped start/end frame, for the 1st time we can control what's in between with the 80+ camera motion and vfx

this is a big level up for AI video

step by step tutorial:

[Dan ⚡️ @d4m1n]

GPT-4o image generation for branding.

this saves days of work 🤯

7 examples + prompts:

[el.cine @EHuanglu]

this new AI is mind blowing..

Stability AI just dropped Stable Virtual Camera

- turn 2D image into 3D video(32 depth)

- control camera movements with 360° spins, spirals, dolly zooms

- keeps everything 3D-consistent for up to 1000 frames

[el.cine @EHuanglu]

this new AI is crazy…

FLORA just dropped a node based AI canvas.

it can analyse your story and generate prompts for your shots, so you can create images and videos directly and..

it can even design characters for your film

step by step tutorial:

[lord asado 🥩 @lordasado]

I made this interactive generative art piece using @Grok 3, purely on my iPhone.

I didn’t write a single line of code. Grok 3 pretty much nailed my idea without a single error.

How is this not AGI?