This model was generated using Meshy(https://www.meshy.ai)'s AI toolkit. It provides tools that effortlessly turn human imagination into stunning 3D models.

If you're interested in this model, visit the model page on Meshy to explore more about it: [https://www.meshy.ai/3d-models/RunningPup-Movers-Trailer-v2-019db1af-7450-779c-b33d-6c7d3dc51006?utm_source=sketchfab](https://www.meshy.ai/3d-models/RunningPup-Movers-Trailer-v2-019db1af-7450-779c-b33d-6c7d3dc51006?utm_source=sketchfab)

RunningPup Movers Trailer

Advanced multipurpose combat mech designed for high mobility warfare, urban operations, and cinematic standing poses.

Optimized for:
• sci-fi scenes
• game concepts
• hard surface rendering
• pretending your GPU isn’t suffering

Mecha 1

SlidingPuzzle

This models contains a detailed PC filled with RGB lights.
BLEND FILE: https://sketchfab.com/3d-models/custom-gaming-pc-1a24273417534f69afa0f7c62b643ffc

Configuration:

i7 10700K

ASUS ROG Maximus XIII Z590 HERO motherboard

G SKILL TRIDENTZ RGB RAM Modules

GALAX RTX 2060 (PREVIEW OF MY UPCOMING GPU PACK)

DEEPCOOL Castle 240 AIO

DEEPCOOL MATTREXX 55 V3 case

Although it is a free model, lot of effort has been put into providing accurate wiring and as many details as possible. Model looks absolutely amazing in blender cycles render.

CUSTOM GAMING PC (FBX file)

This models contains a detailed PC filled with RGB lights.
FBX FILE: https://sketchfab.com/3d-models/custom-gaming-pc-fbx-file-00bb790319c045058290dfbb4873f3ac

Configuration:

i7 10700K

ASUS ROG Maximus XIII Z590 HERO motherboard

G SKILL TRIDENTZ RGB RAM Modules

GALAX RTX 2060 (PREVIEW OF MY UPCOMING GPU PACK)

DEEPCOOL Castle 240 AIO

DEEPCOOL MATTREXX 55 V3 case

Although it is a free model, lot of effort has been put into providing accurate wiring and as many details as possible.
Model looks absolutely amazing in blender cycles render.

Custom Gaming PC

NVLink is a high speed bus from Nvidia that was developed for data transport between RTX GPU nodes and CPUs for example in the area of high performance workstations and rack based supercomputers. In newer computer systems it also replaces the SLI system for the direct connection of graphics cards.

Nvlink For Rtx Graphics Cards

Another idea I decided to put on paper, the Japanese-built Howa Type 20 assault rifle and some custom props by me to make it unique.

I allow a generous tris count for this model, about 102k tris, 4 x 4k texture maps.

Renders: https://www.artstation.com/artwork/280mxx

Loadouts:

- BCM Gunfighter Mod 3 grip, MAGPUL Gen M3, MLOK Paraclip sling swivel & Aluminium Rail & MBUS Sights & AFG 2

- DEON's March-F 1-10×24 FFP Shorty & LaRue M68 CCO mount

- Steiner DBAL I2 Class 1 IR & Surefire M600U Scout Light and SR07-D-IT switch.

*A/N: (Nov. 5th) Texture revision done, added GLX160 grenade launcher.
https://drive.google.com/file/d/1HSwNBQ0hwwAcdZiAvBnv9Fwn2547BmTA/view?usp=sharing
Unfortunately, my Glock 19 is disabled due to copyright. You can view the project here.
https://www.artstation.com/artwork/zOo0qq

Howa Type 20

**3D scan of a mechanical keyboard reconstructed with AliceVision/Meshroom 2025.1.0.**

**Pipeline:** single iPhone 14 Pro 1080p video (orbit, 30 fps, ~60s) → frames @ 4 fps → SAM3 (text prompt "keyboard") for background segmentation → COLMAP-style sparse + dense reconstruction via Meshroom's full `photogrammetry` template → **3D Delaunay tetrahedralization + Graph Cut Max-Flow surface extraction** (instead of Poisson) → native Texturing node bakes a 4096×4096 diffuse map from the masked frames.

**Stats:** 81 frames, 101,432 vertices, 147,881 triangles. Full pipeline ran on a GCE L4 GPU in ~50 minutes.

Keyboard_meshroom_delaunay_sketchfab

This model was generated using Meshy, an AI-powered 3D creation toolkit. Meshy helps creators turn ideas into high-quality 3D models with ease. View the original model and claim a free month of Pro on Meshy: [https://www.meshy.ai/s/hQ9GPU](https://www.meshy.ai/s/hQ9GPU)

The Magpul AFG (Angled Fore Grip) is essentially an angled grip that helps mitigate recoil and control the gun better.

Magpul AFG (Low-Poly)

a low-poly replica of the commando starter weapon of killing floor 2
a lot of sources say this is based off of the colt smg 
in my opinon this is wrong, the rifle has the same damage profile as other 5.56 weapons, the .17 hmr is not as powerful,but the velocity is almost the same as 5.56
RIS foregrip, tan rail covers RRA Tactical Mount rear sight tan Magpul furniture AFG CTR stock
imfdb said that it has a flashlight, I didn't see it.

low-poly Kf2 ar15 "varmint rifle"

A high-fidelity 3D reconstruction of a modern computer laboratory at Daffodil International University. This project focuses on realistic lighting, material accuracy (specifically the brick textures and wooden workstations), and functional space design.

Technical Details:

Software: Autodesk 3ds Max

Renderer: Arnold GPU (Optimized with NVIDIA RTX 5060 Ti)

Key Features: Seamless marble tiling, custom-modeled workstations, and detailed ceiling-mounted AC units.

This model was created as part of a Design Course assignment, achieving a perfect score for its technical execution and lighting balance.

DIU Computer Lab - Realistic Interior ArchViz

Single loop around the transmitter mast

DJI Mini 5 Pro

Images: 293

Weather: Cloudy

Software: WebODM + Blender

Options: auto-boundary:true, dem-resolution:3, dsm:true, feature-quality:ultra, mesh-size:300000, no-gpu:true, orthophoto-resolution:1, pc-quality:ultra

Average GSD: 2 cm

Area: 9,128.12 m²

ReconstructedPoints: 163,544,335

Spatial Reference: GPS

Disk Usage: 	5.18 Gb

Pen-y-Lan Transmitter Mast

**Pipeline:** single iPhone 14 Pro 1080p video (orbit, 30 fps, ~63s) → frames @ 4 fps → SAM3 (text prompt "cup") for background segmentation → COLMAP-style sparse + dense reconstruction via Meshroom's full `photogrammetry` template → **3D Delaunay tetrahedralization + Graph Cut Max-Flow surface extraction** (instead of Poisson) → native Texturing node bakes a 4096×4096 diffuse map from the masked frames.
**Stats:** 252 frames, ~94k vertices, ~118k triangles. Full pipeline ran on a GCE L4 GPU in ~90 minutes.

资源库

关于这些资源