CUSTOMIMAGEOBJECT RECOGNITION
Everyday Object Photo Sets
Everyday object image sets with object IDs, multi-view capture, material/category metadata, angle labels, resolution checks, QA fields, and optional bounding boxes for perception models.
CUSTOMIMAGEPRODUCT RECOGNITION
Product and Packaging Image Sets
Structured product and packaging image collection format for visual product recognition, package understanding, OCR-assisted labeling, retail model evaluation, and catalog enrichment.
Product and Packaging Image Sets are built for teams evaluating how vision models understand consumer products, packaging surfaces, labels, logos, nutrition panels, ingredient blocks, barcodes, and product-side variations across real capture conditions.
Each product entry is structured around a capture set rather than a single image. A standard set can include multiple views of the same item, such as front, back, side, top, angled, label close-up, barcode close-up, or in-context shelf/counter placement where scoped.
Structured collection
Product and Packaging Image Sets are structured around product groups, not isolated photos.
Metadata depth
Supported annotations include product bounding boxes, package panel boxes, text region boxes, barcode region boxes, logo region…
Delivery-ready package
Includes image quality, view coverage, product grouping reviews, plus dataset manifest, CSV catalog, consent status metadata, QA…
Request sample access
Share your target product categories, view requirements, annotation needs, OCR fields, and evaluation goals. We’ll reply with a technical scope for the collection.
Key highlights
- Multi-angle product capture structure with front, back, side, top, angled, and label-focused views.
- Product-level and image-level metadata for category, view angle, background type, lighting condition, resolution, and capture context.
- Optional OCR annotations for visible package text, ingredient blocks, nutrition panels, barcode regions, or label fields.
Dataset contents
Image capture
Product image files · Packaging image files · Product group IDs · View angle labels · Category metadata
Text regions and annotations
Package side metadata · Lighting condition metadata · Background type metadata · Resolution metadata · Capture context metadata
Object and scene metadata
Optional text region boxes · Optional barcode region boxes · Optional OCR transcript fields · Optional product bounding boxes
Technical specifications
Use cases
- Product recognition
- Retail vision model evaluation
- Package understanding
- OCR-assisted product labeling
- Catalog enrichment
- Barcode/text-region detection
- Visual search evaluation
- Shelf and package perception testing
Dataset workflow
1Collection
Product and Packaging Image Sets are structured around product groups, not isolated photos. Contributors capture approved products using defined angle, lighting, background, and label-visibility instructions so each product can be evaluated across multiple visual views.
2Review
Image quality review · View coverage review · Product grouping review · Text visibility review · Annotation consistency review · Consent status review · Delivery approval
3Delivery
Image files · Product metadata CSV · View metadata CSV · OCR files where scoped · Annotation files where scoped · Dataset manifest JSON · CSV file catalog · QA summary · Consent status metadata · Delivery notes
Quality assurance
Related dataset formats
CUSTOMIMAGEOCR
Real-World Text and Signage Photos
Scene-text image sets with language/script metadata, text-region boxes, ground-truth transcription, scene category labels, QA fields, and manifest records for OCR and visual translation.
CUSTOMMULTIMODALPAIRED DATA
Multimodal Paired Capture Sets
Paired multimodal datasets linking video, audio, screen, text, or image-derived records through shared file IDs, cross-modal manifests, metadata schemas, and QA fields.
CUSTOMANY MODALITYBUYER-SCOPED
Buyer-Scoped Custom Collection
Fully scoped custom data collection with buyer-defined modality, schema, metadata fields, annotation layers, QA criteria, manifest structure, delivery package, and collection rules.
CUSTOMVIDEODEXTEROUS TASKS
Hand and Object Interaction Clips
Hand-object interaction clips with action verbs, object categories, camera angle metadata, contact-event markers, object-state labels, and QA fields for robotic manipulation evaluation.
CUSTOMVIDEOENVIRONMENT DIVERSITY
Multi-Environment Activity Clips
Activity video sets captured across varied real-world environments, with task labels, environment metadata, camera position fields, lighting conditions, duration, and QA status for activity recognition and embodied AI evaluation.