We collect a variety of images from all over the world and deliver meaningful results. We have experience in many fields: Downtown, Indoor, Object, Nature, Furniture, Sports, Painting, Animals and plants, Food, Sign.
Quality control for building corpus data:
The Uptempo Data team abides by the ESSENTIAL 5 principles that contain our own know-how. And based on this, we build image data that guarantees optimal quality
Step 1: Raw data collection
-
- Collecting image data in file formats such as JPG and PDF (including text data)
- Filtering and removing data unsuitable for processing and utilization purposes
Step 2: Source data building
-
- Data labelings such as detailed data classification and de-identification of collected raw data
- Standardization and setup with source data in a form that can be processed through crowd working
Step 3: Source data processing
-
- Primary processing: Mark the recognition area of the target image data using a bounding box
- Secondary processing: Input text (label) in the displayed area
Step 4: Processed data inspection
-
- Complete inspection: Implementing basic quality inspection across the construction data by applying strict quality standards
- Cross-checking: Implementation of grouping and cross-quality inspection by applying K-fold cross-validation method
Step 5: Final delivery of construction data
-
- Final delivery to the client only when the final data is of suitable quality
- For the insufficient amount of construction, only high-quality data is delivered by constructing more than 150% of excess data