Debian Codegraph @

Documentation/userspace-api/media/v4l/metafmt-d4xx.rst

raw
1 .. SPDX-License-Identifier: GFDL-1.1-no-invariants-or-later
2 .. c:namespace:: V4L
3
4 .. _v4l2-meta-fmt-d4xx:
5
6 *******************************
7 V4L2_META_FMT_D4XX ('D4XX')
8 *******************************
9
10 Intel D4xx UVC Cameras Metadata
11
12
13 Description
14 ===========
15
16 Intel D4xx (D435, D455 and others) cameras include per-frame metadata in their UVC
17 payload headers, following the Microsoft(R) UVC extension proposal [1_]. That
18 means, that the private D4XX metadata, following the standard UVC header, is
19 organised in blocks. D4XX cameras implement several standard block types,
20 proposed by Microsoft, and several proprietary ones. Supported standard metadata
21 types are MetadataId_CaptureStats (ID 3), MetadataId_CameraExtrinsics (ID 4),
22 and MetadataId_CameraIntrinsics (ID 5). For their description see [1_]. This
23 document describes proprietary metadata types, used by D4xx cameras.
24
25 V4L2_META_FMT_D4XX buffers follow the metadata buffer layout of
26 V4L2_META_FMT_UVC with the only difference, that it also includes proprietary
27 payload header data. D4xx cameras use bulk transfers and only send one payload
28 per frame, therefore their headers cannot be larger than 255 bytes.
29
30 This document implements Intel Configuration version 3 [9_].
31
32 Below are proprietary Microsoft style metadata types, used by D4xx cameras,
33 where all fields are in little endian order:
34
35 .. tabularcolumns:: |p{5.0cm}|p{12.5cm}|
36
37
38 .. flat-table:: D4xx metadata
39 :widths: 1 2
40 :header-rows: 1
41 :stub-columns: 0
42
43 * - **Field**
44 - **Description**
45 * - :cspan:`1` *Depth Control*
46 * - __u32 ID
47 - 0x80000000
48 * - __u32 Size
49 - Size in bytes, include ID (all protocol versions: 60)
50 * - __u32 Version
51 - Version of this structure. The documentation herein covers versions 1,
52 2 and 3. The version number will be incremented when new fields are
53 added.
54 * - __u32 Flags
55 - A bitmask of flags: see [2_] below
56 * - __u32 Gain
57 - Gain value in internal units, same as the V4L2_CID_GAIN control, used to
58 capture the frame
59 * - __u32 Exposure
60 - Exposure time (in microseconds) used to capture the frame
61 * - __u32 Laser power
62 - Power of the laser LED 0-360, used for depth measurement
63 * - __u32 AE mode
64 - 0: manual; 1: automatic exposure
65 * - __u32 Exposure priority
66 - Exposure priority value: 0 - constant frame rate
67 * - __u32 AE ROI left
68 - Left border of the AE Region of Interest (all ROI values are in pixels
69 and lie between 0 and maximum width or height respectively)
70 * - __u32 AE ROI right
71 - Right border of the AE Region of Interest
72 * - __u32 AE ROI top
73 - Top border of the AE Region of Interest
74 * - __u32 AE ROI bottom
75 - Bottom border of the AE Region of Interest
76 * - __u32 Preset
77 - Preset selector value, default: 0, unless changed by the user
78 * - __u8 Emitter mode (v3 only) (__u32 Laser mode for v1) [8_]
79 - 0: off, 1: on, same as __u32 Laser mode for v1
80 * - __u8 RFU byte (v3 only)
81 - Spare byte for future use
82 * - __u16 LED Power (v3 only)
83 - Led power value 0-360 (F416 SKU)
84 * - :cspan:`1` *Capture Timing*
85 * - __u32 ID
86 - 0x80000001
87 * - __u32 Size
88 - Size in bytes, include ID (all protocol versions: 40)
89 * - __u32 Version
90 - Version of this structure. The documentation herein corresponds to
91 version xxx. The version number will be incremented when new fields are
92 added.
93 * - __u32 Flags
94 - A bitmask of flags: see [3_] below
95 * - __u32 Frame counter
96 - Monotonically increasing counter
97 * - __u32 Optical time
98 - Time in microseconds from the beginning of a frame till its middle
99 * - __u32 Readout time
100 - Time, used to read out a frame in microseconds
101 * - __u32 Exposure time
102 - Frame exposure time in microseconds
103 * - __u32 Frame interval
104 - In microseconds = 1000000 / framerate
105 * - __u32 Pipe latency
106 - Time in microseconds from start of frame to data in USB buffer
107 * - :cspan:`1` *Configuration*
108 * - __u32 ID
109 - 0x80000002
110 * - __u32 Size
111 - Size in bytes, include ID (v1:36, v3:40)
112 * - __u32 Version
113 - Version of this structure. The documentation herein corresponds to
114 version xxx. The version number will be incremented when new fields are
115 added.
116 * - __u32 Flags
117 - A bitmask of flags: see [4_] below
118 * - __u8 Hardware type
119 - Camera hardware version [5_]
120 * - __u8 SKU ID
121 - Camera hardware configuration [6_]
122 * - __u32 Cookie
123 - Internal synchronisation
124 * - __u16 Format
125 - Image format code [7_]
126 * - __u16 Width
127 - Width in pixels
128 * - __u16 Height
129 - Height in pixels
130 * - __u16 Framerate
131 - Requested frame rate per second
132 * - __u16 Trigger
133 - Byte 0: bit 0: depth and RGB are synchronised, bit 1: external trigger
134 * - __u16 Calibration count (v3 only)
135 - Calibration counter, see [4_] below
136 * - __u8 GPIO input data (v3 only)
137 - GPIO readout, see [4_] below (Supported from FW 5.12.7.0)
138 * - __u32 Sub-preset info (v3 only)
139 - Sub-preset choice information, see [4_] below
140 * - __u8 reserved (v3 only)
141 - RFU byte.
142
143 .. _1:
144
145 [1] https://docs.microsoft.com/en-us/windows-hardware/drivers/stream/uvc-extensions-1-5
146
147 .. _2:
148
149 [2] Depth Control flags specify which fields are valid: ::
150
151 0x00000001 Gain
152 0x00000002 Exposure
153 0x00000004 Laser power
154 0x00000008 AE mode
155 0x00000010 Exposure priority
156 0x00000020 AE ROI
157 0x00000040 Preset
158 0x00000080 Emitter mode
159 0x00000100 LED Power
160
161 .. _3:
162
163 [3] Capture Timing flags specify which fields are valid: ::
164
165 0x00000001 Frame counter
166 0x00000002 Optical time
167 0x00000004 Readout time
168 0x00000008 Exposure time
169 0x00000010 Frame interval
170 0x00000020 Pipe latency
171
172 .. _4:
173
174 [4] Configuration flags specify which fields are valid: ::
175
176 0x00000001 Hardware type
177 0x00000002 SKU ID
178 0x00000004 Cookie
179 0x00000008 Format
180 0x00000010 Width
181 0x00000020 Height
182 0x00000040 Framerate
183 0x00000080 Trigger
184 0x00000100 Cal count
185 0x00000200 GPIO Input Data
186 0x00000400 Sub-preset Info
187
188 .. _5:
189
190 [5] Camera model: ::
191
192 0 DS5
193 1 IVCAM2
194
195 .. _6:
196
197 [6] 8-bit camera hardware configuration bitfield: ::
198
199 [1:0] depthCamera
200 00: no depth
201 01: standard depth
202 10: wide depth
203 11: reserved
204 [2] depthIsActive - has a laser projector
205 [3] RGB presence
206 [4] Inertial Measurement Unit (IMU) presence
207 [5] projectorType
208 0: HPTG
209 1: Princeton
210 [6] 0: a projector, 1: an LED
211 [7] reserved
212
213 .. _7:
214
215 [7] Image format codes per video streaming interface:
216
217 Depth: ::
218
219 1 Z16
220 2 Z
221
222 Left sensor: ::
223
224 1 Y8
225 2 UYVY
226 3 R8L8
227 4 Calibration
228 5 W10
229
230 Fish Eye sensor: ::
231
232 1 RAW8
233
234 .. _8:
235
236 [8] The "Laser mode" has been replaced in version 3 by three different fields.
237 "Laser" has been renamed to "Emitter" as there are multiple technologies for
238 camera projectors. As we have another field for "Laser Power" we introduced
239 "LED Power" for extra emitter.
240
241 The "Laser mode" __u32 fields has been split into: ::
242 1 __u8 Emitter mode
243 2 __u8 RFU byte
244 3 __u16 LED Power
245
246 This is a change between versions 1 and 3. All versions 1, 2 and 3 are backward
247 compatible with the same data format and they are supported. See [2_] for which
248 attributes are valid.
249
250 .. _9:
251
252 [9] LibRealSense SDK metadata source:
253 https://github.com/IntelRealSense/librealsense/blob/master/src/metadata.h