Ikomia HUB

Inference for MMOCR from MMLAB KIE models

Train for MMOCR from MMLAB KIE models

Inference for MMOCR from MMLAB text recognition models

Inference for MMOCR from MMLAB text detection models

Inference for MMDET from OneDL MMDetection models

Inference for MMLAB segmentation models

Train for MMLAB segmentation models

Train MMDetection models

Run florence 2 segmentation with or without text prompt

Run florence 2 object detection with or without text prompt

Inference for text recognition (OCR) with Florence-2

Image captioning with Florence-2

Inference with D-FINE models

Train D-FINE models

Inference with YOLO26 models (Ultralytics)

Inference with YOLO26 pose estimation models

Train YOLO26 segmentation models.

Train YOLO26 classification models.

Inference with YOLO26 segmentation models

Train YOLO26 object detection models.

Inference with YOLO26 image classification models

Multiple object tracking algorithm for object detection, instance segmentation and keypoints

Multiple Object Tracking algorithm (MOT) combining a deep association metricwith the well known SORT algorithm for better performance.

Infer ByteTrack for object tracking

Face analysis with detection, age, gender, and emotion prediction using UniFace

Face parsing (semantic segmentation) using UniFace BiSeNet model.

Face verification using UniFace - verify if two faces belong to the same person

Face search using UniFace - find matching faces in an image database

Lightweight production-ready face analysis library built on ONNX Runtime.

Inference for Segment Anything Model 3 (SAM3) - Interactive instance segmentation.