Model Tutorials

Intel OpenVINO supports most of the TensorFlow and PyTorch models. The table below lists some deep learning models that commonly used in the Embodied Intelligence solutions. You can find information about how to run them on Intel platforms:

Algorithm	Description	Link
YOLOv8	CNN based object detection	https://github.com/openvinotoolkit/openvino_notebooks/tree/latest/notebooks/yolov8-optimization
YOLOv12	CNN based object detection	https://github.com/openvinotoolkit/openvino_notebooks/tree/latest/notebooks/yolov12-optimization
MobileNetV2	CNN based object detection	https://github.com/openvinotoolkit/open_model_zoo/blob/master/models/public/mobilenet-v2-1.0-224
SAM	Transformer based segmentation	https://github.com/openvinotoolkit/openvino_notebooks/tree/latest/notebooks/segment-anything
SAM2	Extend SAM to video segmentation and object tracking with cross attention to memory	https://github.com/openvinotoolkit/openvino_notebooks/tree/latest/notebooks/sam2-image-segmentation
FastSAM	Lightweight substitute to SAM	https://github.com/openvinotoolkit/openvino_notebooks/tree/latest/notebooks/fast-segment-anything
MobileSAM	Lightweight substitute to SAM (Same model architecture with SAM. Can refer to OpenVINO SAM tutorials for model export and application)	https://github.com/openvinotoolkit/openvino_notebooks/tree/latest/notebooks/segment-anything
U-NET	CNN based segmentation and diffusion model	https://community.intel.com/t5/Blogs/Products-and-Solutions/Healthcare/Optimizing-Brain-Tumor-Segmentation-BTS-U-Net-model-using-Intel/post/1399037?wapkw=U-Net
DETR	Transformer based object detection	https://github.com/openvinotoolkit/open_model_zoo/tree/master/models/public/detr-resnet50
GroundingDino	Transformer based object detection	https://github.com/openvinotoolkit/openvino_notebooks/tree/latest/notebooks/grounded-segment-anything
CLIP	Transformer based image classification	https://github.com/openvinotoolkit/openvino_notebooks/tree/latest/notebooks/clip-zero-shot-image-classification
Qwen2.5VL	Multimodal large language model	https://github.com/openvinotoolkit/openvino_notebooks/tree/latest/notebooks/qwen2.5-vl
Whisper	Automatic speech recognition	https://github.com/openvinotoolkit/openvino_notebooks/tree/latest/notebooks/whisper-asr-genai
FunASR	Automatic speech recognition	Refer to the FunASR Setup in LLM Robotics sample pipeline

Attention

When following these tutorials for model conversion, please ensure that the OpenVINO version used for model conversion is the same as the runtime version used for inference. Otherwise, unexpected errors may occur, especially if the model is converted using a newer version and the runtime is an older version. See more details in the Troubleshooting.

Please also find information for the models of imitation learning, grasp generation, simultaneous localization and mapping (SLAM) and bird’s-eye view (BEV):

Note

Before using these models, please ensure that you have read the AI Content Disclaimer.

• Action Chunking with Transformers - ACT
• Visual Servoing - CNS
• Diffusion Policy
• Improved 3D Diffusion Policy (iDP3)
• Feature Extraction Model: SuperPoint
• Feature Tracking Model: LightGlue
• Bird’s Eye View Perception: Fast-BEV
• Monocular Depth Estimation: Depth Anything V2
• Robotics Diffusion Transformer (RDT-1B)