GstInference with InceptionV4 layer

GstInference
Introduction
Getting started
Getting the code Building the plugin Quick Starting Guide
Supported architectures
InceptionV1 InceptionV2 InceptionV3 InceptionV4 TinyYoloV2 TinyYoloV3 YoloV2 ResNetV1 MobileNetV2 MobileNetV2 + SSD AlexNet
Supported backends
TensorFlow Tensorflow-Lite Coral from Google Caffe TensorRT ONNXRT ONNXRT ACL ONNXRT OpenVINO
Metadata and Signals
GstClassificationMeta GstEmbeddingMeta GstDetectionMeta GstInferenceMeta Signals
Overlay Elements
Inference Overlay
Utils Elements
InferenceFilter InferenceCrop InferenceDebug InferenceBin
Legacy pipelines
PC NANO TX2 Xavier IMX8
Example pipelines
Example applications
Classification Detection Hand Pose Recognition
Benchmarks
Model Zoo
Project Status
Releases Roadmap
Contact Us

Description

Inception v4 is a deep convolutional network architecture that has been shown to achieve very good performance at a relatively low computational cost. Its performance is similar to the latest generation Inception-v3 network, but it adds residual connections in conjunction with more traditional architecture.

Architecture

The network uses Inception v3 as a base but incorporates residual connections.

Inception v4 architecture

Gstreamer Plugin

The GStreamer plugin uses the pre-process and post-process described in the original paper. Please take into consideration that not all deep neural networks are trained the same even if they use the same model architecture. If the model is trained differently, details like label ordering, input dimensions, and color normalization can change.

This element was based on the ncappzoo repo. The pre-trained model used to test the element may be downloaded from our R2I Model Zoo for the different frameworks.

Pre-process

Input parameters:

• Input size: 229 x 229
• RGB Mean: [128, 128, 128]
• RGB Standard deviation: [128, 128, 128]
• Format RGB

The pre-process consists of taking the input image and transforming it to the input size (by scaling, interpolation, cropping...). Then the mean is subtracted from each pixel on RGB, and it is divided by the standard deviation. No, no color space conversion is needed because the model was trained with RGB.

Post-process

The model output is a float array of size 1000 containing the probability for each one of the ImageNet labels. The post-process consists of simply searching the highest probability on the array.

Examples

Please refer to the Googlenet section on the examples page.