DriveWorks SDK Reference
3.5.78 Release
For Test and Development only

perception/landmarks/camera/roadmarknet/docs/roadmarknet_mainsection.md
Go to the documentation of this file.
1 # Copyright (c) 2020 NVIDIA CORPORATION. All rights reserved.
2 
3 @page roadmarknet_mainsection RoadMarkNet
4 
5 ## About This Module
6 This module provides the APIs to initialize, query, and release the NVIDIA proprietary road mark classification deep neural network: **RoadMarkNet**. <br>
7 
8 The data structures include:
9  - `::dwRoadMarkNetModel`: allows user to choose among RoadMarkNet models.
10  - `::dwRoadMarkNetParams`: in addition to `::dwRoadMarkNetModel`, defines the RoadMarkNet model variant with specific precision and processor optimization to be loaded.
11 
12 DriveWorks comes with the following RoadMarkNet model:
13  - US only model: ::DW_ROADMARKNET_MODEL_US_V0: Supported on GPU
14 
15 ### Inputs
16 
17 RoadMarkNet APIs consume FP16 planar frames with a resolution of 960x604 pixels from AR0231 cameras (revision >= 4) and dwLandmarkDetection from LandMarkDetector. Internally, it uses road mark bounding box information from dwLandmarkDetection to crop input images and resize them to an image patch of 64x64. This image patch is then fed to the DNN as an input.
18 
19 ### Outputs
20 
21 RoadMarkNet currently supports 33 road mark classes in US:
22  - BackGround
23  - ArrowMarkStraight
24  - ArrowMarkLeft
25  - ArrowMarkRight
26  - ArrowMarkStraightLeft
27  - ArrowMarkStraightRight
28  - ArrowMarkLeftRight
29  - ArrowMarkStraightLeftRight
30  - ArrowMarkMerge
31  - SpeedLimitMark5MPH
32  - SpeedLimitMark10MPH
33  - SpeedLimitMark15MPH
34  - SpeedLimitMark20MPH
35  - SpeedLimitMark25MPH
36  - SpeedLimitMark30MPH
37  - SpeedLimitMark35MPH
38  - SpeedLimitMark40MPH
39  - SpeedLimitMark45MPH
40  - SpeedLimitMark50MPH
41  - SpeedLimitMark55MPH
42  - SpeedLimitMark60MPH
43  - SpeedLimitMark65MPH
44  - SpeedLimitMark70MPH
45  - SpeedLimitMark75MPH
46  - SpeedLimitMark80MPH
47  - TextMarkBike
48  - TextMarkAhead
49  - TextMarkStop
50  - TextMarkSchool
51  - TextMarkHump
52  - BikeMark
53  - HOVMark
54  - HumpMark
55 
56 ### Additional Information
57 @warning The RoadMarkNet model currently deployed is mostly trained with United States road markings. It might experience reduced accuracy in other regions, particularly for road marking patterns not present in the US.
58 
59 ## Relevant Tutorials
60 
61 - @ref roadmarknet_usecase1
62 
63 ## APIs
64 
65 - @ref roadmarknet_group