PeopleNet AMR Model Card

Model Overview

The Peoplenet Autonomous Mobile Robot (AMR) model described in this card detect one or more physical objects from people category within an image, predicts a box around each object, and returns a masked image aggregating all boxes around detected objects. The model is based on PeopleNet, and optimized for people detection on RealSense applications running at 480x640 image resolution.

Model Architecture

This model is based on NVIDIA DetectNet_v2 detector with ResNet34 as feature extractor. This architecture, also known as GridBox object detection, uses bounding-box regression on a uniform grid on the input image. Gridbox system divides an input image into a grid which predicts four normalized bounding-box parameters (xc, yc, w, h) and confidence value per output class.

The raw normalized bounding-box and confidence detections are the postprocessed to a masked image, where all pixels within bounding boxes are masked with 1.

Training

This model was trained using the DetectNet_v2 entrypoint in TAO. The training algorithm optimizes the network to minimize the localization and confidence loss for the objects. The training is carried out in two phases. In the first phase, the network is trained without regularization.

Training Data

PeopleNet AMR model was trained on a proprietary dataset with more than 3 million images and more than 8 million objects for person class. The training dataset consists of a mix of camera heights, crowd-density, and field-of view (FOV). Approximately two thirds of the training data consisted of images captured in a indoor and outdoor environments that are from a horizontal view point. For this case, the camera is typically set up at approximately 2 to 5 feet height, 90-degree angle and has wide field-of-view. We have also added approximately 45 thousand images with low-density scenes from a robot's point of view to improve the performance for use-cases where person object detection is needed at low heights.

Training Data Ground-truth Labeling Guidelines

The training dataset is created by labeling ground-truth bounding-boxes and categories by human labellers. Following guidelines were used while labelling the training data for NVIDIA PeopleNet AMR model. If you are looking to re-train with your own dataset, please follow the guideline below for highest accuracy.

PeopleNet AMR project labelling guidelines:

1. All objects that fall under person in the image and are larger than the smallest bounding-box limit for the corresponding class (height >= 10px OR width >= 10px @1920x1080) are labeled with the appropriate class label.
2. If a person is carrying an object please mark the bounding-box to include the carried object as long as it doesn’t affect the silhouette of the person. For example, exclude a rolling bag if they are pulling it behind them and are distinctly visible as separate object. But include a backpack, purse etc. that do not alter the silhouette of the pedestrian significantly.
3. Occlusion: For partially occluded objects that do not belong a person class and are visible approximately 60% or are marked as visible objects with bounding box around visible part of the object. These objects are marked as partially occluded. Objects under 60% visibility are not annotated.
4. Occlusion for person class: If an occluded person’s head and shoulders are visible and the visible height is approximately 20% or more, then these objects are marked by the bounding box around the visible part of the person object. If the head and shoulders are not visible please follow the Occlusion guidelines in item 3 above.
5. Truncation: For an object other than a person that is at the edge of the frame with visibility of 60% or more visible are marked with the truncation flag for the object.
6. Truncation for person class: If a truncated person’s head and shoulders are visible and the visible height is approximately 20% or more mark the bounding box around the visible part of the person object. If the head and shoulders are not visible please follow the Truncation guidelines in item 5 above.
7. Each frame is not required to have an object.

Performance

Evaluation Data

The inference performance of PeopleNet AMR v0.1 model was measured against more than 15000 proprietary images across a variety of environments. The frames are high resolution images 1920x1080 pixels resized to 960x544 pixels before passing to the PeopleNet AMR detection model.

Methodology and KPI

The true positives, false positives, false negatives are calculated using intersection-over-union (IOU) criterion greater than 0.5. The KPI for the evaluation data are reported in the table below. The FP16 Model is evaluated based on precision, recall and accuracy.

How to use this model

This model needs to be used with NVIDIA Hardware and Software. For Hardware, the models can run on any NVIDIA GPU including NVIDIA Jetson devices. These models can only be used with TensorRT.

The primary use case intended for this model is detecting people in a color (RGB) image. The model can be used to detect people from photos and videos by using appropriate video or image decoding and pre-processing.

Input

RGB Image of dimensions: 480 x 640 X 3 (H x W x C). Channel Ordering of the Input: NHWC, where N = Batch Size, H = Height of images (480), W = Width of the images (640) , C = number of channels (3). Input scale: 1/255.0 Mean subtraction: None

Output

Masked image : 480 x 640 x 1 (H x W x C) where pixels indicating detected people are 1. and 0. otherwise. Channel Ordering of the Output: NHWC, where N = Batch Size, H = Height of images (480), W = Width of the images (640) , C = number of channels (1). Input scale: 0. / 1.

Limitations

Very Small Objects

NVIDIA PeopleNet AMR model were trained to detect objects larger than 10x10 pixels. Therefore it may not be able to detect objects that are smaller than 10x10 pixels.

Occluded Objects

When objects are occluded or truncated such that less than 20% of the object is visible, they may not be detected by the PeopleNet AMR model. For people class objects, the model will detect occluded people as long as head and shoulders are visible. However if the person’s head and/or shoulders are not visible, the object might not be detected unless more than 60% of the person is visible.

Dark-lighting, Monochrome or Infrared Camera Images

The PeopleNet AMR model were trained on RGB images in good lighting conditions. Therefore, images captured in dark lighting conditions or a monochrome image or IR camera image may not provide good detection results.

Warped and Blurry Images

The PeopleNet AMR models were not trained on fish-eye lense cameras. Therefore, the models may not perform well for warped images and images that have a lot of blur.

Face and Bag class

Although bag and face class are included in the model, the accuracy of these classes will be much lower than people class. Some re-training will be required on these classes to improve accuracy.

References

Citations

• Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You only look once: Unified, real-time object detection. In: CVPR. (2016)
• Erhan, D., Szegedy, C., Toshev, A., Anguelov, D.: Scalable object detection using deep neural networks, In: CVPR. (2014)
• He, K., Zhang, X., Ren, S., Sun, J.: Deep Residual Learning for Image Recognition. In: CVPR (2015)

License

License to use these models is covered by the Model EULA. By downloading the unpruned or pruned version of the model, you accept the terms and conditions of these licenses.

Ethical Considerations

NVIDIA PeopleNet AMR model detects faces. However, no additional information such as race, gender, and skin type about the faces is inferred.

Training and evaluation dataset mostly consists of North American content. An ideal training and evaluation dataset would additionally include content from other geographies.

NVIDIA’s platforms and application frameworks enable developers to build a wide array of AI applications. Consider potential algorithmic bias when choosing or creating the models being deployed. Work with the model’s developer to ensure that it meets the requirements for the relevant industry and use case; that the necessary instruction and documentation are provided to understand error rates, confidence intervals, and results; and that the model is being used under the conditions and in the manner intended.