これは個人のUE5学習プロジェクトです。 These are personal UE5 learning projects.
TOKYO
- Tweets 951
- Following 224
- Followers 11,082
- Likes 3,490
If you want to go fast, go alone. If you want to go far, go together.
If you care about games, building games, making art, 3D or making content being accessible to everyone, you've probably seen that I've been obsessed with making the best open source 3D and asset generation pipeline and I've been blazing past checkpoints. But I don't think anyone can do it alone. 2600+ clones in 2 weeks, contributors raising PRs and issues means that this is something that many people care about. And everyone can contribute!
The pipeline is free and open source - so test it, share it, break it - but most of all talk about it and tell us what you'd like to see and what you don't like....
The Algo doesn't care much about text posts, so I'm sharing the github repo - because interacting with the post tells the algo you care. So if you see this - help out in any way you can. Every input, opinion, like, comment, bookmark, and share counts.
I have to get back to my day-job - but I'll see you in the replies, on discord or github. We have a bunch of things rolling out in the next few days that make your 3D characters even better.
github.com/Bingeljell/image-…
Free image-to-3D models don't preserve your source image - text, logos and fine detail come back garbled.
I got that working with my latest 'Pixel Match' feature in my image-to-3dlab: it copies the actual pixels from your source onto the model, even after retopo down to ~5k faces. High quality, low poly 3D models that are riggable and can be animated.
One step closer to Tripo / Meshy level of output, but free and on your own machine. Mac + NVIDIA.
Try it out here: github.com/Bingeljell/image-…
There's lots more to do, so please test, raise issues & PRs, etc.
6
10
130
8,062
I tried using the previous version!
Here is my feedback: During the process of reducing the high-poly model, it would be great if we could set an upper limit on the polygon count. Also, if the reduction precision/quality improves to the point where models can be directly used in games, I think the user base will grow significantly. Currently, the mesh structure still requires manual cleanup/adjustments before it can be implemented into a game.
I’ll definitely test out the rigging and animation once those adjustments are refined, but I think what you're building is truly amazing. Keep up the great work, I'm cheering you on!
1
3
230
#UE5
LocalFogVolumeすごいねー…砂塵… 見てるだけで目が痒くなってきた🥹
デバッグにPrintStringを多用してたけど、
デバッグメニュー作って全部出してしまえば（脳筋）良いなって。
場所覚えられればそこそこありかも。
16
99
953
23,564
¡La dispersión de luz en la tormenta de arena quedó impecable! Totalmente de acuerdo en retirar los PrintString.
1
272
私の翻訳が正しいことを願っています。素晴らしいアイデアがあります。各プレイヤーが操作するモンスターを、さまざまなパーツや外見で強化・カスタマイズできるオンラインゲームにしてみてはどうでしょうか？そうすれば、他のプレイヤーとオンラインで対戦できます。きっと壮大なゲームになるでしょ
1
172
Its been a legitimate dream of mine for the last 10+ years if I somehow had the financial backing to breathe life into the next gen MMORPG ZOIDS game that not only would deliver on every level the fans deserve but would pull in newcomers...id love to contribute even as an idea...
1
1
27
What if you used image generators for the mecha design?
You could probably ask the AI you connected to unreal to make some mecha designs that are inspired by zoids in the web browser chat.
You could even sketch something then have it generate a full image.
1
85
Opus5.5 × Godot
せっかくClaudeがアクションゲームを作ってくれたので、剣のエフェクトも作ってみました⚔️✨
作り方は解説動画に🎥
色々書いてますがこのままOpusに見せれば作ってくれると思います🤣
@godotengine
Opus5.5 × Krea × Godot
Blenderを使わず、イラストから3Dモデル制作✨
①GPT Imageでイラスト生成🖼️
②Kreaの「3D Objects」にあるTRELLIS 2で3Dモデル化🧸
③Opus5.5がGodot用に、原点の位置を足元に合わせ、サイズや質感も調整。骨入れ（リギング）までやってくれました😆
Codex（GPT Image）もKreaもGodotも、すべてClaudeに連携できるので、Claudeとのチャットだけで制作が完結‼️
ちなみに、アクションゲームを作るつもりはなかったです。
3枚目の指示を送ったら、Claudeが勝手にアクションゲームを作ってきました😦
@krea_ai
20
67
867
74,094
The desert looks very realistic. I'm more impressed by the running animation - feels very natural. The walk still feels a little off, but the run is really good.
Great job!
1
151
