libvisiontransfer  6.1.0
README.md
1 Vision Transfer Library 6.1.0
2 -----------------------------
3 
4 The given library provides functionally for receiving and transmitting
5 image pairs over a computer network. The intended use for this library
6 is to receive output data from Nerian's SceneScan sensor systems.
7 However, the library also provides functionality for transmitting image
8 pairs. It can thus be used for emulating SceneScan when performing
9 systems development, or for transmitting image data to SceneScan when
10 using network image input.
11 
12 Images can be transferred with a bit depth of either 8 or 12 bits. When
13 receiving 12-bit images, the library inflates the images internally to
14 16 bits, in order to allow for more efficient processing. Monochrome
15 and RGB images are supported.
16 
17 When receiving data from SceneScan, the first image of an image pair is
18 typically the rectified image of the left camera, with a bit depth of 8
19 or 12 bits. The second image is typically a disparity map with subpixel
20 resolution, which is transmitted as a 12-bit image. Each value in the
21 disparity map has to be divided by 16 in order to receive disparities
22 at the correct scale.
23 
24 There exist three possible ways for receiving and transmitting image
25 pairs:
26 
27 * `ImageProtocol` is the most low-level interface. This class allows for the
28  encoding and decoding of image pairs to / from network messages.
29  You will have to handle all network communication yourself.
30 
31 * `ImageTransfer` opens up a network socket for sending and receiving
32  image pairs. This class is single-threaded and will thus block when
33  receiving or transmitting data.
34 
35 * `AsyncTransfer` allows for the asynchronous reception or transmission of
36  image pairs. This class creates one or more threads that handle all
37  network communication.
38 
39 In order to discover connected devices on the network, the class
40 `DeviceEnumeration` can be used, which scans for available devices and
41 returns a list of `DeviceInfo` objects.
42 
43 Further, the library includes the class `Reconstruct3D`, which can be used
44 for transforming a received disparity map into a set of 3D points.
45 
46 [Changelog](CHANGELOG.md)
Nerian Vision Technologies