Dimension Collapse

Class: DimensionCollapseBlockV1

Source: inference.core.workflows.core_steps.fusion.dimension_collapse.v1.DimensionCollapseBlockV1

Takes multiple step outputs at data depth level n, concatenate them into list and reduce dimensionality to level n-1.

Useful in scenarios like: * aggregation of classification results for dynamically cropped images * aggregation of OCR results for dynamically cropped images

Type identifier

Use the following identifier in step "type" field: roboflow_core/dimension_collapse@v1to add the block as as step in your workflow.

Properties

Name	Type	Description	Refs
name	str	Enter a unique identifier for this step..	❌

The Refs column marks possibility to parametrise the property with dynamic values available in workflow runtime. See Bindings for more info.

Available Connections

Compatible Blocks

Check what blocks you can connect to Dimension Collapse in version v1.

• inputs: Expression, Time in Zone, OpenAI, Line Counter, Line Counter Visualization, Cosine Similarity, SIFT Comparison, Dimension Collapse, LMM, Qwen2.5-VL, Data Aggregator, Multi-Label Classification Model, Model Comparison Visualization, Line Counter, OCR Model, Stability AI Image Generation, Image Slicer, QR Code Detection, Distance Measurement, Detection Offset, VLM as Detector, Keypoint Detection Model, Keypoint Detection Model, Delta Filter, YOLO-World Model, Slack Notification, Label Visualization, Byte Tracker, Image Threshold, Local File Sink, Cache Set, Polygon Visualization, Byte Tracker, Identify Outliers, Dominant Color, VLM as Classifier, Buffer, Pixelate Visualization, SIFT, Environment Secrets Store, Instance Segmentation Model, Detections Consensus, Clip Comparison, Rate Limiter, Bounding Rectangle, Keypoint Visualization, Polygon Zone Visualization, Template Matching, Stitch OCR Detections, Detections Stabilizer, Stability AI Inpainting, OpenAI, QR Code Generator, Stability AI Outpainting, Detections Stitch, Multi-Label Classification Model, Time in Zone, SIFT Comparison, Roboflow Dataset Upload, Blur Visualization, Velocity, Identify Changes, Color Visualization, Instance Segmentation Model, SmolVLM2, Halo Visualization, Mask Visualization, Time in Zone, Property Definition, Path Deviation, Background Color Visualization, Perception Encoder Embedding Model, VLM as Classifier, Byte Tracker, Detections Filter, Trace Visualization, Google Vision OCR, Dot Visualization, Classification Label Visualization, Model Monitoring Inference Aggregator, Stitch Images, Anthropic Claude, Email Notification, Pixel Color Count, Absolute Static Crop, Object Detection Model, Image Contours, Dynamic Crop, CSV Formatter, LMM For Classification, JSON Parser, Gaze Detection, CogVLM, Object Detection Model, Image Preprocessing, Circle Visualization, CLIP Embedding Model, Florence-2 Model, PTZ Tracking (ONVIF).md), Relative Static Crop, Image Slicer, Moondream2, Roboflow Dataset Upload, Cache Get, Reference Path Visualization, Depth Estimation, Dynamic Zone, Bounding Box Visualization, Icon Visualization, Roboflow Custom Metadata, Detections Transformation, Florence-2 Model, Path Deviation, Ellipse Visualization, Grid Visualization, Camera Focus, Single-Label Classification Model, Image Blur, Detections Classes Replacement, First Non Empty Or Default, Continue If, Size Measurement, Crop Visualization, Corner Visualization, Triangle Visualization, Twilio SMS Notification, Clip Comparison, Webhook Sink, OpenAI, Overlap Filter, Camera Calibration, Image Convert Grayscale, Single-Label Classification Model, Llama 3.2 Vision, Segment Anything 2 Model, VLM as Detector, Barcode Detection, Google Gemini, Perspective Correction, Detections Merge
• outputs: Reference Path Visualization, Path Deviation, Bounding Box Visualization, Cache Set, VLM as Classifier, Florence-2 Model, Polygon Visualization, Time in Zone, Path Deviation, Trace Visualization, VLM as Classifier, Buffer, Ellipse Visualization, OpenAI, Grid Visualization, Dot Visualization, Instance Segmentation Model, Classification Label Visualization, Detections Consensus, Clip Comparison, Line Counter, Line Counter Visualization, Email Notification, Anthropic Claude, Keypoint Visualization, Polygon Zone Visualization, Size Measurement, Triangle Visualization, Corner Visualization, Crop Visualization, Line Counter, Object Detection Model, LMM For Classification, Clip Comparison, Webhook Sink, OpenAI, Object Detection Model, Time in Zone, Llama 3.2 Vision, Circle Visualization, Roboflow Dataset Upload, VLM as Detector, Florence-2 Model, VLM as Detector, Keypoint Detection Model, Instance Segmentation Model, Keypoint Detection Model, Color Visualization, Halo Visualization, Google Gemini, Perspective Correction, Mask Visualization, YOLO-World Model, Time in Zone, Label Visualization, Roboflow Dataset Upload

Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds Dimension Collapse in version v1 has.

Bindings

• input

  • data (*): Reference to step outputs at depth level n to be concatenated and moved into level n-1..

• output

  • output (list_of_values): List of values of any type.

Example JSON definition of step Dimension Collapse in version v1

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/dimension_collapse@v1",
    "data": "$steps.ocr_step.results"
}