のた

オリジナルツイートは殆ど消滅

[はぅ君 @Hau_kun]

t=0
draw=_=>{t++||createCanvas(W=720,W)
background(0)
noFill()
stroke(W)
circle(360,360,30)
for(a=-1;a<2;a+=2)for(r=0;r<6+!(D=27)+!(X=Y=360);r+=PI/3)for(d=0;d<350;d+=(D-=1))line(X,Y,X=360+cos(R=r+noise(d/W-t/W)*6*a)*d,Y=360+sin(R)*d)+circle(X,Y,D)}
#つぶやきProcessing

[CGbox @vtuberkaibou]

BlenderのGeometry Nodesだけで、無限に歩き回れるような「バックルーム」風の空間を作る大型チュートリアル
https://t.co/Lufg9Eqf7S

【主な内容】
・グリッド、カーブ変換、ノイズ判定を使い、モデリングやUV展開なしで迷路状の壁と通路を自動生成する
・壁紙、床、天井素材をボックス投影で割り当て、長い廊下や壁の開口部もGeometry Nodes内で追加
・蛍光灯の点灯量をスライダーで制御し、色補正、グレア、ブルーム、フィルム風の粒を加えて不気味な映像感に仕上げる #b3d

[Miettinen Jesse - Embrace FX @JesseMiettinen]

Just an idea for local voronoi mapping. Would need multiple hole types for sure! #b3d #shadermetrynodes

[Akiyoshi Kitaoka @AkiyoshiKitaoka]

扇形は同じ場所にいるわけではない。

[AssetSaleLove @汗人柱 @AssetLove]

【新着無料アセット】
かわいい空が53種類も入ってるの嬉しすぎる…😳💙
​
爽やかな青空から、夢みたいな夕焼け、幻想的な夜空にオーロラまで入った、アニメ風のスカイボックス集
​
空を変えるだけで、ワールドの雰囲気かなり盛れる☁️✨
​
『Midgard Skybox Prime』(無料)
https://t.co/3szbyVSJvt
#Unity ゲーム開発 VRChat ワールド制作
By.MaNa

👇注目アセットをピックアップ！フレッシュセール開催中
https://t.co/CrnqcaRPP8

[Mike Murdock @mjmurdoc]

Sped up the ships and added some camera tilt
#shader #ui #space

[apek @apekshik]

exploring a design system based on finite element analysis

[Kosta Derpanis @CSProfKGD]

Visualization of the day: Steerable filters

[TenMomo @TenMomo260104]

Unityでリアルタイム濡れ表現💧

標準のLitマテリアルを維持したままWetを適用できるようになりました🙌
内部ではCustomFunctionでShaderGraphに値を持っていく構造ですが、Litは設定済みSGに差し替えるだけで値を全て維持したまま濡れ表現が追加されます。
これで何でも濡らすことができます😌

今はジオメトリ形状だけで計算してますが、いずれHeightMapに対応して、マップ凹凸に沿って流れるようになる予定🌊

#gamedev #indiedev #unity

[真あつきatsuki @atsuki17173305]

今回のライブでも使用するいい感じのメタボールパーティクル🫧

[ashlee! @ashlee3dee]

nobody:

basalt columns when you aren't looking

[Xor @XorDev]

Fever
f z,d
@(70)
{
f3 p = z * nor(2*C.rgb - R.xyy)
p.xy *= mat2(cos(z*.5+f4(,33,11,)))
p.z-=T;
d=2; @(5) d+=d,
p += sin(p.yzx*d+z) / d
z += min(abs(cos(p.y)),d=len(1/tan(p.xz)))/4;
O += f4(1.1+sin(p),)/d
}
O = tanh(O / 2e2)

[Mathelirium @mathelirium]

I love turning basic Math concepts into art.

This is a Riemann sphere driven by a moving rational function

R(z) = ∏(z-aⱼ) / ∏(z-bₖ)

The glowing roots and poles orbit across S² while log|R| and Φ = arg R continuously reshape the surface. The small particles ride the phase gradient ∇Φ.

[Ryohei Sasaki@engineer @rsasaki0109]

ARDY: Autoregressive Diffusion with Hybrid Representation for Interactive Human Motion Generation
SIGGRAPH 2026
https://t.co/SaG6OnMwpX
ARDY is an autoregressive diffusion model designed for interactive motion generation, supporting online text prompting and flexible long-horizon kinematic constraints (root paths/waypoints, full-body keyframes, and sparse joint positions/rotations) with real-time responsiveness.

[Davide Scaramuzza @davsca1]

Check out our #ECCV2026 paper "Low-latency Event-based Object Detection with Spatially-Sparse Linear Attention", where we make linear attention sparse in space, recurrent in time, and parallel in training, enabling the first purely-linear-attention-based neural network for asynchronous object detection with #EventCameras, outperforming the previous best asynchronous method with 20x less computation with truly event-by-event inference on CPU! Code released!

Paper: https://t.co/5eLPgR9kWx
Code: https://t.co/3sNKrWJbVw
Video: https://t.co/6HL3mYOpFL

Event cameras promise extremely low-latency vision, but to fully exploit them, the neural network must be low-latency too. We introduce #SpatiallySparseLinearAttention (#SSLA) for asynchronous object detection directly from raw events.

Linear attention is particularly appealing for event cameras: it can be trained efficiently in parallel on long event sequences, while at inference it operates recurrently, updating its prediction every time a new event arrives.

The problem is that conventional linear attention updates its entire state for every event. For object detection, where fine spatial resolution matters, this quickly becomes expensive.

Our key idea is simple: an event only carries information about a small spatial region, so why update the entire spatial state? SSLA updates only the relevant parts of the state, enabling fine-grained spatial representations while keeping per-event computation low.
We achieve:
- >20× lower per-event computation than the strongest prior asynchronous baseline
- State-of-the-art accuracy among asynchronous object detection methods
- Truly event-by-event inference on CPU, designed to preserve the latency advantage of event cameras

Come to our poster on Friday September 11, 2026 from 4-6pm at ExHall #389

Reference:
Haiqing Hao, Zhipeng Sui, Rong Zou, Zijia Dai, Nikola Zubić, Davide Scaramuzza, Wenhui Wang
Low-latency Event-based Object Detection with Spatially-Sparse Linear Attention
ECCV, 2026

@Prophesee_ai @SynSenseNeuro @UZH_en
@UZH_Science @ERC_Research @UZH_ai @uzh_ifi @Tesla @BYDCompany

#EventCameras #ComputerVision #Robotics #DeepLearning #NeuromorphicVision #AI

[VinciEye @vincieye]

Getting a relightable 3D mesh from photos used to mean messy topology, no UVs, and baked-in shadows. ExMesh++ fixes that: adaptive topology + PBR materials from multi-view images. Clean assets, no ambiguity ⚡

📄 https://t.co/SbG8FTwBQr

[Ryohei Sasaki@engineer @rsasaki0109]

LightGlue ONNX
ONNX-compatible LightGlue: Local Feature Matching at Light Speed. Supports TensorRT, OpenVINO
https://t.co/Q1vDiknqVP
Open Neural Network Exchange (ONNX) compatible implementation of LightGlue: Local Feature Matching at Light Speed. The ONNX model format allows for interoperability across different platforms with support for multiple execution providers, and removes Python-specific dependencies such as PyTorch. Supports TensorRT and OpenVINO.

[Yadong Xie @yadong_xie]

最近一直在探索 agent 借助 web 技术来表达场景交互的界限在哪里

最后得到的答案是没有界限

光棱坦克，启动！

[Carlito Jr. Calixtro @Carlix1229]

Some purple explosion #2dfx 💥😈