KadirErturk/jenny-tts-tags-6h-v1
Viewer
•
Updated
•
4k
•
31
Kadir Erturk
KadirErturk
·
AI & ML interests
sensors and machine learning
Recent Activity
new activity
3 days ago
marduk-ra/F5-TTS-Turkish:Any plan to change licensing to MIT or Apache 2.0?
updated
a dataset
20 days ago
KadirErturk/jenny-tts-tags-6h-v1
reacted
to
singhsidhukuldeep's
post
with 🔥
about 1 month ago
Good folks at @nvidia and @Tsinghua_Uni have released LLAMA-MESH - A Revolutionary Approach to 3D Content Generation!
This innovative framework enables the direct generation of 3D meshes from natural language prompts while maintaining strong language capabilities.
Here is the Architecture & Implementation!
>> Core Components
Model Foundation
- If you haven't guessed it yet, it's built on the LLaMA-3.1-8B-Instruct base model
- Maintains original language capabilities while adding 3D generation
- Context length is set to 8,000 tokens
3D Representation Strategy
- Uses the OBJ file format for mesh representation
- Quantizes vertex coordinates into 64 discrete bins per axis
- Sorts vertices by z-y-x coordinates, from lowest to highest
- Sorts faces by the lowest vertex indices for consistency
Data Processing Pipeline
- Filters meshes to a maximum of 500 faces for computational efficiency
- Applies random rotations (0°, 90°, 180°, 270°) for data augmentation
- Generates ~125k mesh variations from 31k base meshes
- Uses Cap3D-generated captions for text descriptions
>> Training Framework
Dataset Composition
- 40% Mesh Generation tasks
- 20% Mesh Understanding tasks
- 40% General Conversation (UltraChat dataset)
- 8x training turns for generation, 4x for understanding
Training Configuration
- Deployed on 32 A100 GPUs (for Nvidia, this is literally in-house)
- 21,000 training iterations
- Global batch size: 128
- AdamW optimizer with a 1e-5 learning rate
- 30-step warmup with cosine scheduling
- Total training time: approximately 3 days (based on the paper)
This research opens exciting possibilities for intuitive 3D content creation through natural language interaction. The future of digital design is conversational!
Organizations
None yet
KadirErturk's activity
Any plan to change licensing to MIT or Apache 2.0?
3
#10 opened 4 days ago
by
KadirErturk
updated
a
dataset
20 days ago
reacted to
singhsidhukuldeep's
post with 🔥
about 1 month ago
Post
2299
Good folks at
@nvidia
and @Tsinghua_Uni have released LLAMA-MESH - A Revolutionary Approach to 3D Content Generation!
This innovative framework enables the direct generation of 3D meshes from natural language prompts while maintaining strong language capabilities.
Here is the Architecture & Implementation!
>> Core Components
Model Foundation
- If you haven't guessed it yet, it's built on the LLaMA-3.1-8B-Instruct base model
- Maintains original language capabilities while adding 3D generation
- Context length is set to 8,000 tokens
3D Representation Strategy
- Uses the OBJ file format for mesh representation
- Quantizes vertex coordinates into 64 discrete bins per axis
- Sorts vertices by z-y-x coordinates, from lowest to highest
- Sorts faces by the lowest vertex indices for consistency
Data Processing Pipeline
- Filters meshes to a maximum of 500 faces for computational efficiency
- Applies random rotations (0°, 90°, 180°, 270°) for data augmentation
- Generates ~125k mesh variations from 31k base meshes
- Uses Cap3D-generated captions for text descriptions
>> Training Framework
Dataset Composition
- 40% Mesh Generation tasks
- 20% Mesh Understanding tasks
- 40% General Conversation (UltraChat dataset)
- 8x training turns for generation, 4x for understanding
Training Configuration
- Deployed on 32 A100 GPUs (for Nvidia, this is literally in-house)
- 21,000 training iterations
- Global batch size: 128
- AdamW optimizer with a 1e-5 learning rate
- 30-step warmup with cosine scheduling
- Total training time: approximately 3 days (based on the paper)
This research opens exciting possibilities for intuitive 3D content creation through natural language interaction. The future of digital design is conversational!
This innovative framework enables the direct generation of 3D meshes from natural language prompts while maintaining strong language capabilities.
Here is the Architecture & Implementation!
>> Core Components
Model Foundation
- If you haven't guessed it yet, it's built on the LLaMA-3.1-8B-Instruct base model
- Maintains original language capabilities while adding 3D generation
- Context length is set to 8,000 tokens
3D Representation Strategy
- Uses the OBJ file format for mesh representation
- Quantizes vertex coordinates into 64 discrete bins per axis
- Sorts vertices by z-y-x coordinates, from lowest to highest
- Sorts faces by the lowest vertex indices for consistency
Data Processing Pipeline
- Filters meshes to a maximum of 500 faces for computational efficiency
- Applies random rotations (0°, 90°, 180°, 270°) for data augmentation
- Generates ~125k mesh variations from 31k base meshes
- Uses Cap3D-generated captions for text descriptions
>> Training Framework
Dataset Composition
- 40% Mesh Generation tasks
- 20% Mesh Understanding tasks
- 40% General Conversation (UltraChat dataset)
- 8x training turns for generation, 4x for understanding
Training Configuration
- Deployed on 32 A100 GPUs (for Nvidia, this is literally in-house)
- 21,000 training iterations
- Global batch size: 128
- AdamW optimizer with a 1e-5 learning rate
- 30-step warmup with cosine scheduling
- Total training time: approximately 3 days (based on the paper)
This research opens exciting possibilities for intuitive 3D content creation through natural language interaction. The future of digital design is conversational!
liked
a
model
3 months ago
updated
a
model
10 months ago
liked
a
model
10 months ago
liked
a
model
11 months ago
Unable to run the example code on ubuntu
1
#8 opened over 1 year ago
by
KadirErturk