Generates synthetic datasets for training and evaluating vision models on object permanence and occlusion tracking tasks. Each sample contains objects that are temporarily blocked by a moving horizontal mask, testing the ability to track objects through occlusion.
Each sample pairs a task (first frame + prompt describing what needs to happen) with its ground truth solution (final frame showing the result + video demonstrating how to achieve it). This structure enables both model evaluation and training.
| Property | Value |
|---|---|
| Task ID | G-36 |
| Task | Multiple Occlusions Horizontal |
| Category | Transformation |
| Resolution | 1024×1024 px |
| FPS | 16 fps |
| Duration | ~2 seconds |
| Output | PNG images + MP4 video |
# 1. Clone the repository
git clone https://github.com/VBVR-DataFactory/G-36_multiple_occlusions_horizontal_data-generator.git
cd G-36_multiple_occlusions_horizontal_data-generator
# 2. Create and activate virtual environment
python3 -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# 3. Install dependencies
pip install --upgrade pip
pip install -r requirements.txt
pip install -e .# Generate 50 samples
python examples/generate.py --num-samples 50
# Custom output directory
python examples/generate.py --num-samples 100 --output data/my_dataset
# Reproducible generation with seed
python examples/generate.py --num-samples 50 --seed 42
# Without videos (faster)
python examples/generate.py --num-samples 50 --no-videos| Argument | Description |
|---|---|
--num-samples |
Number of tasks to generate (required) |
--output |
Output directory (default: data/questions) |
--seed |
Random seed for reproducibility |
--no-videos |
Skip video generation (images only) |
The scene shows 4 objects arranged horizontally on the right side of the frame, with a dark rectangular mask initially positioned on the left side. Move the mask horizontally to the right in a continuous motion until it leaves the frame. As it moves, the mask passes in front of the objects, temporarily blocking them from view.
 |
 |
 |
| Initial Frame Mask on left, objects visible on right |
Animation Mask moves right, occluding objects sequentially |
Final Frame Mask exits frame, objects remain visible |
Move a horizontal mask across stationary objects, temporarily occluding them as it passes, then continue until the mask leaves the frame.
- Objects: 3 shapes arranged horizontally on right side
- Mask: Dark rectangular occluder starting on left side
- Movement: Mask moves horizontally from left to right
- Occlusion effect: Objects disappear behind mask, reappear after
- Objects remain stationary: Only mask moves
- Background: White with clear contrast
- Goal: Complete horizontal traversal with sequential occlusion
- Horizontal occlusion dynamics
- Multiple sequential occlusions (3 objects)
- Object permanence testing (objects persist behind mask)
- Continuous smooth mask motion
- Objects revealed again after occlusion
- Tests tracking through temporary blocking
data/questions/multiple_occlusions_horizontal_task/multiple_occlusions_horizontal_00000000/
├── first_frame.png # Mask on left, objects visible on right
├── final_frame.png # Mask exited, objects visible again
├── prompt.txt # Horizontal occlusion instruction
├── ground_truth.mp4 # Animation of mask moving across objects
└── question_metadata.json # Task metadata (objects, positions, etc.)
File specifications:
- Images: 1024×1024 PNG format
- Video: MP4 format, 16 fps
- Duration: ~2 seconds
visual-reasoning occlusion object-permanence tracking horizontal-motion sequential-occlusion