OpenFilter Hub
Powered by OpenFilter, A Universal Abstraction for Computer Vision Workloads
Hyperlabel
1.0.0CommunityModels
Download HyperlabelDownloadable AI annotation and model training software. Automated annotation of video, images, audio, medical imaging, satellite imagery and other datasets.
Image Streamer
1.5.1CommunityConnectors (Video/Streaming)
Download Image StreamerWrite a images from video streams to disk
SUSE AI Filter
0.1.0CommercialModels
SUSE AI Compliant Distribution of OpenFilter
Plate ID OCR
0.1.16CommunityUtility
Download Plate ID OCRDemo Application: Overlays subject data from DriveID, Clipper, and TextScan on processed license plate frames
DEIMv2 Detection
0.3.0CommercialModels
Real-time object detection powered by the DEIMv2 (DETR with Improved Matching, v2) architecture. Runs multi-topic detection with configurable NMS (standard/weighted/soft), result aggregation across frames, per-class ignore lists, and optional overlay visualization.
Color Attribute
0.6.0CommercialModels
Tags the color of each detected object (cars, clothing, people) so footage can be searched by color. Sits after an object detector, reads the color of every detection box, and writes color and color_margin onto each detection; with no detector it reads the whole frame. Supports a siglip2 vision-language backend or classical hsv/rgb pixel voting, an optional viz topic, and per-part clothing colors via human parsing.
Scene Caption
0.6.1CommercialModels
Vision-language scene captioning. Buffers video frames in a rolling window and runs a VLM (Gemini cloud or Cosmos-Reason2-2B local) every N seconds to produce natural-language scene descriptions. Supports frames or video input mode (Gemini), Vertex AI auth, detection-gated inference, and single-object focus mode.
Tracking ReID
0.1.14CommercialModels
Appearance-based re-identification embedding extraction for tracked objects
Pose Estimation
0.1.14CommercialModels
Human pose estimation using MediaPipe or RTMPose backends
RT-DETR
0.1.21CommercialModels
Real-time object detection using RT-DETR architecture with built-in IoU tracking
Global Association Multicamera Tracking
0.1.13CommercialModels
Cross-camera identity association using ReID embeddings, temporal gating, and configurable transition maps
Huggingface Vision
0.4.17CommunityModels
Download Huggingface VisionGeneric vision filter using Hugging Face Transformers for object detection and image classification across video streams
Event Sink
1.1.11CommunityConnectors (Streams & Queues)
Download Event SinkA production-ready OpenFilter component that collects events from filter pipelines and reliably delivers them to a CloudEvents ingestion compatible endpoint.
HyperLabel (SAM3 Detector)
0.1.33CommunityModels
Download HyperLabel (SAM3 Detector)Segment Anything Model 3 for object detection and segmentation. This filter is a part of Plainsight's HyperLabel™, which enables accelerated/automated annotation, model retraining and deployment promotion.
Protege Model
0.1.33CommercialModels
Trainable model to run on Plainsight
Video Source
1.5.1CommunityConnectors (Video/Streaming)
Download Video SourceRead a single video file, network stream or webcam until the end
Video Streamer
1.5.1CommunityConnectors (Video/Streaming)
Download Video StreamerWrite a single segmented video file or rtsp stream
Visual Calibration
2.0.10CommunityUtility
Download Visual CalibrationPerforms frame calibration, shake detection
Clipper
0.1.22CommunityUtility
Download Clippercrops any region of interest (polygon) in a video and creates a new video from it
Web Viewer
1.5.1CommunityConnectors (Video/Streaming)
Download Web ViewerShow incoming topic image stream on a web server
ImageIn
1.5.1CommunityConnectors (Storage & Files)
Download ImageInRead a images from a folder or GCS bucket or S3 bucket
Aggregator
1.1.11CommunityUtility
Download AggregatorAggregates metadata into combined results (summing stats ..etc)
HyperLabel (ChatTag)
0.3.7CommunityModels
Download HyperLabel (ChatTag)Label images using ChatGPT. This filter is a part of Plainsight's HyperLabel™, which enables accelerated/automated annotation, model retraining and deployment promotion.
REST Connect
1.5.1CommunityConnectors (Streams & Queues)
Download REST ConnectProvide REST endpoint(s) and send incoming JSON data on down the filter pipeline
FaceGuard
1.4.6CommunityUtility
Download FaceGuardAutomatically detects and blurs faces in video streams.
FrameSelect
1.3.6CommunityUtility
Download FrameSelectDetect motion and structural changes, and save video frames that meet specific criteria
GCS Upload
2.0.10CommunityUtility
Download GCS UploadUploads video segments and images to Google Cloud Storage buckets.
Quality Gate
1.2.8CommercialUtility (Testing)
Benchmarks and evaluates model performance.
Data Capture
1.5.1CommunityConnectors (Databases & Search)
Download Data CaptureRecord frame data to files as JSON or CSV. If the output filename ends with '.csv' the output will be CSV, otherwise JSON
MQTT Bridge
1.5.1CommunityConnectors (Streams & Queues)
Download MQTT BridgeExports data to MQTT, bundled in with OpenFilter