Loading video...

Video Failed to Load

Go Home

Explore the color-coded male torso blockout from every angle on 3D REF: This 3D model clearly illustrates each muscle's location, base forms, and how the back muscles overlap. #anatomy #characterdesign #3dmodeling

134,925 views • 1 year ago •via X (Twitter)

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

I really like those 2.5D characters here's my Seedance2 prompt: MANGA READER TURNAROUND CHARACTER: Young Japanese man sitting cross-legged on the floor reading a colorful manga magazine, spiky messy black hair sticking up in chunky angular shapes, thick eyebrows, small round glasses, slight stubble, cross pendant necklace on a thin chain, wearing a loose open dark navy pinstripe yukata robe showing his chest, bare feet tucked under his legs, small ceramic ashtray with cigarette butts on the floor to his left, round ceramic pot to his right, focused slightly annoyed expression looking down at the magazine. 3D CG model with 2D hand-drawn overlay, scribble hatching in all shadow areas boiling every frame, thick ink outlines on silhouette edges redrawing with wobble, flat painted color textures on 3D geometry, color bleeding outside contour lines. SEQUENCE [0s–5s] Character sits cross-legged holding the manga open in both hands, not moving, just reading. Camera begins a smooth continuous 360 degree orbit around him at his eye level. As the camera moves from front to three-quarter to side profile, the 3D model rotates cleanly revealing his full volume but the 2D ink layer on top lives and breathes, scribble hatching in the shadows under his jaw and inside the yukata folds and between his crossed legs redraws every frame with jittery boiling energy, thick ink outlines on his silhouette wobble and shift in thickness as the angle changes, the flat painted navy yukata wraps around his torso showing new pinstripe ink lines and fold marks at each new angle, his spiky hair reveals its 3D chunky volume from new angles, the manga magazine pages catch the light differently as the camera passes, cigarette smoke curls upward as loose scribbled white lines drifting gently [5s–10s] Camera continues the smooth unbroken orbit past his side revealing the cross necklace hanging from his neck in profile, through his back showing the yukata fabric draped across his shoulders and the spiky hair from behind, past his left side and returning to front, the entire rotation fluid and steady, throughout the full 360 the boiling ink outlines never stop trembling on every edge, the scribble hatching in shadows keeps redrawing in slightly different positions each frame giving the whole image a living sketchbook vibration even though he holds perfectly still reading, the flat painted color planes on his skin shift between warm peach and darker shadow tones as new surfaces catch the light, the ashtray and ceramic pot on the floor rotate into and out of view naturally, white background STYLE: 2.5D painterly CG. 3D CG geometry underneath for volume and smooth rotation. 2D Grease Pencil layer on top, thick wobbly ink outlines, scribble hatching in all shadows boiling every frame, flat painted color planes not smooth gradients. No photorealism. No smooth CG rendering. Muted palette dark navy warm peach black white. Film grain. Clean white background. Smooth continuous camera orbit.

INK

15,192 views • 2 months ago

Wonderland: Navigating 3D Scenes from a Single Image Contributions: • First, we introduce a representation for controllable 3D generation by leveraging the generative priors from camera-guided video diffusion models. Unlike image models, video diffusion models are trained on extensive video datasets. This enables them to capture comprehensive spatial relationships within scenes across multiple views and embed a form of "3D awareness" in their latent space, which allows us to maintain 3D consistency in novel view synthesis. • Second, to achieve controllable novel view generation, we empower video models with precise control over specified camera motions. We introduce a novel dual-branch conditioning mechanism that effectively incorporates desired diverse camera trajectories into the video diffusion model. This enables expansion of a single image into a multi-view consistent capture of a 3D scene with precise pose control. • Third, to achieve efficient 3D reconstruction, we directly transform video latents into 3DGS. We propose a novel latent-based large reconstruction model (LaLRM) that lifts video latents to 3D in a feed-forward manner. With this design, during inference, our model directly predicts 3DGS from a single input image, effectively aligning the generation and reconstruction tasks—and bridging image space and 3D space—through the video latent space. Compared with reconstructing scenes from images, the video latent space offers a 256× spatial-temporal reduction while retaining essential and consistent 3D structural details. Such a high degree of compression is crucial, as it allows the LaLRM to handle a wider range of 3D scenes within the reconstruction framework, with the same memory constraints.

MrNeRF

52,849 views • 1 year ago