Generates synthetic datasets for training and evaluating vision models on pattern recognition and sequential reasoning tasks. Each sample contains a small-large alternating size sequence where the next figure must be selected from multiple choices.

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.

# 1. Clone the repository
git clone https://github.com/VBVR-DataFactory/G-135_select_next_figure_small_large_alternating_sequence_data-generator.git
cd G-135_select_next_figure_small_large_alternating_sequence_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

A sequence of shapes arranged in a 'small-big-small' pattern. Circle the next shape in the candidate area that continues this 'small-big-small-big' pattern.

Initial Frame Small-large alternating sequence and candidates	Animation Correct option identified and circled	Final Frame Answer circled in candidate area

Identify the small-large alternating size pattern in a sequence and select the correct next figure from multiple choices.

• Sequence area: Shapes alternating between small and large
• Candidate area: Multiple options for next figure
• Pattern constraints: Same shape, same color, alternating size
• Alternation pattern: Small → Large → Small → Large (starting with small)
• Shapes: 10 types (square, circle, triangle, diamond, pentagon, hexagon, star, octagon, heart, arrow)
• Colors: 30 distinct colors (red, blue, green, yellow, purple, orange, pink, cyan, lime, indigo, teal, amber, rose, violet, emerald, sky, fuchsia, coral, mint, navy, maroon, olive, turquoise, salmon, lavender, crimson, gold, silver, bronze, plum)
• Task: Identify alternation pattern, find matching option, circle it
• Background: White with clear area separation
• Goal: Select the one option that continues the alternation correctly

• Size alternation pattern recognition (small-to-large focus)
• Binary small/large classification
• Multiple choice selection task
• Tests understanding of alternating sequences
• Shape and color consistency verification
• Visual IQ test style reasoning

data/questions/select_next_figure_small_large_alternating_sequence_task/select_next_figure_small_large_alternating_sequence_00000000/
├── first_frame.png      # Sequence and candidates without answer
├── final_frame.png      # Correct choice circled
├── prompt.txt           # Alternation pattern continuation instruction
├── ground_truth.mp4     # Animation of answer selection
└── question_metadata.json # Task metadata

File specifications:

• Images: 1024×1024 PNG format
• Video: MP4 format, 16 fps
• Duration: ~3-4 seconds

visual-reasoning pattern-recognition size-alternation sequential-reasoning multiple-choice iq-test-style