PP-LiteSeg | Luxonis

Our new model ZOO works with DepthAI V3. Find out more in our documentation.

0 Likes

Model Details

Model Description

PP-LiteSeg: a superior real-time semantic segmentation model for ADAS (advanced driving-assistance systems). The model can segment 19 different classes including cars, pedestrians, traffic signs, bicycles, roads, trees, buildings, and so on.

Developed by: Baidu Inc.
Shared by:
Model type: Computer vision
License:
Resources for more information:

Training Details

Training Data

The model was trained on which is a large-scale dataset for urban segmentation. It contains 5,000 fine annotated images, which are further split into 2975, 500, and 1525 images for training. For more information about training data check the .

Testing Details

Metrics

Model was evaluated on .

Metric	Value
mIoU (%)	72.00 %

Results are taken from .

Technical Specifications

Input/Output Details

Input:
- Name: x
  - Info: NCHW BGR 0-255 image.
Output:
- Name: bilinear_interp_v2_13.tmp_0
  - Info: Segmentation masks for every class.

Model Architecture

It consists of three modules:

Encoder: Lightweight network
Aggregation: Simple Pyramid Pooling Module (SPPM)
Decoder: Flexible and Lightweight Decoder (FLD) and Unified Attention Fusion Module (UAFM)

Please consult the for more information on model architecture.

Throughput

Model variant: pp-liteseg:1024x512

• Input shape: [1, 3, 512, 1024] • Output shape: [1, 19, 512, 1024]

• GFLOPs: 12.213

Platform	Precision	Throughput (infs/sec)	Power Consumption (W)
RVC2	FP16	5.76	N/A
RVC4	INT8	164.08	5.03

* Benchmarked with , using 2 threads (and the DSP runtime in balanced mode for RVC4).

* Parameters and FLOPs are obtained from the package.

Quantization

RVC4 version of the model was quantized using a HubAI Driving dataset.

Utilization

Models converted for RVC Platforms can be used for inference on OAK devices. DepthAI pipelines are used to define the information flow linking the device, inference model, and the output parser (as defined in model head(s)). Below, we present the most crucial utilization steps for the particular model. Please consult the docs for more information.

Install DAIv3 and depthai-nodes libraries:

pip install depthai
pip install depthai-nodes

Define model:

model_description = dai.NNModelDescription(
    "luxonis/pp-liteseg:1024x512"
)

nn = pipeline.create(ParsingNeuralNetwork).build(
    <CameraNode>, model_description
)

Inspect model head(s):

SegmentationParser that outputs message (segmentation mask for 19 classes).

Get parsed output(s):

while pipeline.isRuning():
    parser_output: dai.ImgFrame = parser_output_queue.get()

Example

You can quickly run the model using our script. It automatically downloads the model, creates a DepthAI pipeline, runs the inference, and displays the results using our DepthAI visualizer tool. To try it out, run:

python3 main.py \
    --model luxonis/pp-liteseg:1024x512 \
    -overlay

Semantic segmentation model for advanced driving-assistance systems (ADAS).
License	Apache 2.0 Commercial use
Downloads	173
Tasks	Semantic Segmentation
Model Types	ONNX

Model Variants

Name	Version	Available For	Created At	Deploy
		RVC2, RVC4	10 months ago