What's Changed

• Moved Replay requirements to optional requirements by @Erol444 in https://github.com/luxonis/depthai/pull/757
• [Calibration] Graceful handling when frames cannot be retrieved by @themarpe in https://github.com/luxonis/depthai/pull/762
• Fixed launcher tag parsing to not cause "Unknown" version even after updating: 50671637b69c682991153db81f7bdc2013aecd2b

Full Changelog: https://github.com/luxonis/depthai/compare/v3.1.1...v3.1.2

What's Changed

• Include App in all cases by @moratom in https://github.com/luxonis/depthai/pull/752

Full Changelog: https://github.com/luxonis/depthai/compare/v3.1.0...v3.1.1

What's Changed

• Set default median filter to 5x5 by @szabi-luxonis in https://github.com/luxonis/depthai/pull/706
• Eeprom v7 flash update in calibration by @saching13 in https://github.com/luxonis/depthai/pull/707
• OAK-D-W board json file by @saching13 in https://github.com/luxonis/depthai/pull/721
• Modified charuco board so it's 2.2cm per square on 15" screen by @Erol444 in https://github.com/luxonis/depthai/pull/722
• Revert SDK NNetManager draw API by @szabi-luxonis in https://github.com/luxonis/depthai/pull/727
• Record-replay support in SDK and depthai_demo by @Erol444 in https://github.com/luxonis/depthai/pull/728
• Sdk update 1.2.2 by @Erol444 in https://github.com/luxonis/depthai/pull/732
• Fixed depthai_demo - createDepth from SDK and args path location by @Erol444 in https://github.com/luxonis/depthai/pull/733
• depthai 2.17.0.0 - supports new S2/Pro devices, OV9782 as RGB/center camera by @alex-luxonis in https://github.com/luxonis/depthai/pull/737
• depthai_demo: add -tun/--cameraTuning option by @alex-luxonis in https://github.com/luxonis/depthai/pull/741
• Fixed args parser by @Erol444 in https://github.com/luxonis/depthai/pull/742
• Add initial dynamic recalibration by @szabi-luxonis in https://github.com/luxonis/depthai/pull/690
• [Windows Installer Fix] Catch if missing requirements by @moratom in https://github.com/luxonis/depthai/pull/746
• [Windows Installer Fix] Updated ps1 script by @Erol444 in https://github.com/luxonis/depthai/pull/744

New Contributors

• @moratom made their first contribution in https://github.com/luxonis/depthai/pull/746

Full Changelog: https://github.com/luxonis/depthai/compare/v3.0.12...v3.1.0

What's Changed

• Enable RGB alignment by default for more accurate depth measurement
• Added timer for image capture when calibrating device
• Update depthai to 2.15.4.0

Full Changelog: https://github.com/luxonis/depthai/compare/v3.0.11...v3.0.12

What's Changed

• Fix depthai-sdk broken url link by @ibaiGorordo in https://github.com/luxonis/depthai/pull/649
• Fixed rgb measured translation by @saching13 in https://github.com/luxonis/depthai/pull/656
• Added 'debug' command line option by @themarpe in https://github.com/luxonis/depthai/pull/661
• Updated board files and names by @saching13 in https://github.com/luxonis/depthai/pull/662
• Update depthai to latest release 2.15.1.0 by @alex-luxonis in https://github.com/luxonis/depthai/pull/667
• Added launcher fix for multiple tags on same commit by @themarpe in https://github.com/luxonis/depthai/pull/670
• Add temperature shutdown by @VanDavv in https://github.com/luxonis/depthai/pull/665
• enable extended disparity switch and fix missing device issue by @VanDavv in https://github.com/luxonis/depthai/pull/671
• Sdk tutorial by @ZigiMigi in https://github.com/luxonis/depthai/pull/659
• Model fixes by @VanDavv in https://github.com/luxonis/depthai/pull/675
• Added readme shields by @Erol444 in https://github.com/luxonis/depthai/pull/676
• Update depthai to 2.15.2.0 by @szabi-luxonis in https://github.com/luxonis/depthai/pull/677

New Contributors

• @ibaiGorordo made their first contribution in https://github.com/luxonis/depthai/pull/649
• @ZigiMigi made their first contribution in https://github.com/luxonis/depthai/pull/659

Full Changelog: https://github.com/luxonis/depthai/compare/v3.0.10...v3.0.11

In this release, we added support for OAK-D-Pro, together with GUI configuration options to adjust the IR projector

In this release, we increased the performance of demo to 30 fps, and fixed some regression BUGs from v3.0.8.

In this release, we increased the performance on RPi to >10 FPS.

We're still working on more improvements, that will be available in the next releases

UVC (webcam) app fixed to run on Windows.

In this release we addressed the windows installer issue, failing if a user requested an installation path that contained spaces. With this release, these paths should be handled correctly

What's Changed

• Add ColorOrder property to createColorCam in pipeline manager by @tersekmatija in https://github.com/luxonis/depthai/pull/584
• Version bump SDK to 1.1.6 by @tersekmatija in https://github.com/luxonis/depthai/pull/586
• add fix for Windows paths by @VanDavv in https://github.com/luxonis/depthai/pull/587
• exclude latest version of Qt due to Windows issue by @VanDavv in https://github.com/luxonis/depthai/pull/578
• Apps by @VanDavv in https://github.com/luxonis/depthai/pull/583
• Add sentry integration by @VanDavv in https://github.com/luxonis/depthai/pull/585
• Add ability to run apps from GUI by @VanDavv in https://github.com/luxonis/depthai/pull/589
• remove emotions recognition network (temporarily) by @VanDavv in https://github.com/luxonis/depthai/pull/590
• Add keyboard interrupt catch blocks by @VanDavv in https://github.com/luxonis/depthai/pull/592
• Release tweaks by @VanDavv in https://github.com/luxonis/depthai/pull/593

Full Changelog: https://github.com/luxonis/depthai/compare/v3.0.4...v3.0.5

In this release, we added some improvements to our Qt GUI, increasing both reliability of the script, improving UX and introducing "failsafe" mode. The major change to the code is the new supervisor mode, enabled by default, that is responsible for setting up the environment before actually running the demo. It will also determine if Qt5 is installed on your system or if it crashes curing the runtime - if one of these occur, the demo will fallback to "failsafe" mode and use OpenCV to show the results.

Nice thing noticeable in this release is the loading procedure being now more descriptive, showing e.g. download progress, as on the image below

Full changelog

Features

• Introduce supervisor mode by @VanDavv in https://github.com/luxonis/depthai/pull/573
• Add download progress indicator in Qt GUI by @VanDavv in https://github.com/luxonis/depthai/pull/577

Bugs

• disable disparity encoding to prevent OOM issue by @VanDavv in https://github.com/luxonis/depthai/pull/569
• Remove rectified streams detections mirroring by @VanDavv in https://github.com/luxonis/depthai/pull/574
• Ubuntu 18.04 fix by @VanDavv in https://github.com/luxonis/depthai/pull/571

Misc

• Increase wait time for the demo subprocess to finish by @VanDavv in https://github.com/luxonis/depthai/pull/565

Full Changelog: https://github.com/luxonis/depthai/compare/v3.0.3...v3.0.4

Features

• Add statistics aggregation by @VanDavv in https://github.com/luxonis/depthai/pull/531
• Added capability to force update to a new version by @themarpe in https://github.com/luxonis/depthai/pull/540
• Installer Windows Signing by @themarpe in https://github.com/luxonis/depthai/pull/539
• Migrate QT to pyQt5 by @VanDavv in https://github.com/luxonis/depthai/pull/534
• Allow to specify which render backend to use (-gt / --guiType) by @VanDavv in https://github.com/luxonis/depthai/pull/550
• draw FPS in PreviewManager by @VanDavv in https://github.com/luxonis/depthai/pull/563
• Add human-pose-estimation-0001 support by @VanDavv in https://github.com/luxonis/depthai/pull/566

Bugs

• Add a workaround for GUI blank window on Windows: by @Erol444 in https://github.com/luxonis/depthai/pull/532
• Fix installing requirements on Windows: pyrebase -> pyrebase4 by @alex-luxonis in https://github.com/luxonis/depthai/pull/551
• fix broken example usages links in SDK docs by @VanDavv in https://github.com/luxonis/depthai/pull/555
• Add DEPTHAI_INSTALL_SIGNAL_HANDLER=0 to avoid segfault on RPi by @VanDavv in https://github.com/luxonis/depthai/pull/558

Misc

• Fix VideoEncoder deprecated warnings by @alex-luxonis in https://github.com/luxonis/depthai/pull/543
• Update to latest depthai: 2.13.3.0 by @alex-luxonis in https://github.com/luxonis/depthai/pull/561
• Install numpy version based on python version by @szabi-luxonis in https://github.com/luxonis/depthai/pull/560

Full Changelog: https://github.com/luxonis/depthai/compare/v3.0.2...v3.0.3

With this release, OAK-D-Lite support is added. The demo script is able to recognize the device as any other regular device, no extra steps required

Full Changelog: https://github.com/luxonis/depthai/compare/v3.0.1...v3.0.2

In this release, our demo got a bunch of bug fixes for issues we discovered after the initial 3.0.0 release

• Fixed switching between OAK-1 and OAK-D devices
• Fixed config object passing between threads
• Fixed callbacks, so that using custom callback file works again
• Removed encoders from rectifiedLeft and rectifiedRight streams to prevent OOM issue on OAK-D-PoE

Up till this point, the DepthAI demo script required a terminal to be run, and all the controls and features were hidden behind command line arguments.

With this release, we bring a new look at the demo script, introducing a QT GUI

Now, you can browse through all of the demo features with ease and modify them with a few mouse clicks.

Moreover, to ease the install process, we added a Windows installer (an .exe file) that performs all necessary setup and launches a new demo automatically.

Update depthai dependency to 2.10.0.0

Last release before DepthAI SDK

Final release for Gen1 codebase.

Fix for model downloader when there is space in path. Moved open3d install as optional. Added -fusb2 for calibration. Switched to 400p stereo resolution by default. Added orientation check in calibration. Flip camera orientation for OAK-1 by default. Removed deprecated CLI options. Increased manual exposure limits.

Included changes in the release:

[global] - Ability to manually control focus and exposure of the RGB camera

[depthai-python] - Wheels are prebuilt for Python 3.9, remove 3.5 support due to its EOL

[depthai-core] - Fix watchdog recreation after device object deletion - Add the ability to fetch USB speed and MyriadX serial number

Hi all! Yesterday we released a new DepthAI version - 0.3.0.0!

Changelog:

• Added color, rectified_left and rectitied_right streams (what is rectification?), together with an ability to run a NN on those streams
• Added median_kernel_size and lr_check options to depth pipeline config (you can use -med and -lrc flags in demo script to test them out)
• Added verbose output (-vv) to demo script which will print packets' metadata
• Added a Point Cloud Visualization (use -s rectified_right depth -pcl flags in demo script to preview)
• Added model downloader and removed all NN blobs from depthai repository - they are now downloaded when requested
• Fixed missing metadata for disparity, disparity_color and depth streams
• Fixed sequence number sync issue by adding a sync_sequence_numbers config option (you can test it by using the -seq flag with the demo script)
• Improved calibration system (ability to use two homography)
• Improved format of NN config file

Migrate from the previous release

The major change is related to the new NN config file format. There are now two inference modes: detection and raw, and you can skip the config file entirely, which will default to raw mode.

While in the raw mode, the DepthAI won't attempt to decode nn output tensor and send it directly to host, where you'll have to parse it manually. As an example, take a look how age gender recognition handler has changed: ```diff - detections = [] - for _, e in enumerate(nnet_packet.entries()):
- if e[1]["female"] > 0.8 or e[1]["male"] > 0.8: - detections.append(e[0]["age"])
- if e[1]["female"] > e[1]["male"]:
- detections.append("female")
- else:
- detections.append("male")
- return detections

• output_list = nnet_packet.getOutputsList()
• age_out = output_list[0]
• age = age_out[0,0,0,0]
• gender_out = output_list[1]
• female_conf = gender_out[0,0,0,0]
• male_conf = gender_out[0,1,0,0] +
• detection = None +
• conf_thr = 0.5
• if female_conf > conf_thr or male_conf > conf_thr:
• gender = "male"
• if female_conf > male_conf:
• gender = "female"
• age = (int)(age*100)
• detection = dict(gender=gender, age=age)
• return detection ```

While it looks more complex, it gives more freedom in using NN with custom outputs.

The other mode, detection, requires a config file to be in one of the following forms: - MobilenetSSD-based - YOLO-based

When using this mode, you can use the getDetectedObjects method, which will produce a Detection object

diff - entries_prev = [] - for e in nnet_packet.entries(): - if e[0]['id'] == -1.0 or e[0]['confidence'] == 0.0: - break - if e[0]['confidence'] > 0.5: - entries_prev.append(e[0]) + detections = list(nnet_packet.getDetectedObjects())

Next, you'll need to update places where you referenced the detection object to properly use the new format, just as the following

```diff - for i, e in enumerate(entries_prev): - left = int(e['left'] * img_w) - top = int(e['top'] * img_h) - right = int(e['right'] * img_w) - bottom = int(e['bottom'] * img_h)

• for detection in detections:
• left = int(detection.x_min * img_w)
• top = int(detection.y_min * img_h)
• right = int(detection.x_max * img_w)
• bottom = int(detection.y_max * img_h) ```

First DepthAI Release

Depthai documentation: https://docs.luxonis.com/

Release Notes:

• Calibration Improvements including checking for synchronization and displaying UI stacked.
• Cleanup of example scripts to allow running w/ -cnn option (see here for example usage).
• Fix example scripts to use the latest technique for using w/ and w/out depth (see examples, here).
• Enable configurable max distance threshold (configurable here)
• Option to enable bounding box on depth stream (use the -bb option).
• Timeout for sending neural results if grayscale cameras are not detected (e.g. if they are disconnected intentionally or unintentionally).
• Ability to run H.264 and/or H.265 encoding on color sensor alone. (See how to use here)
• Reduce resolution off of IMX378 to save resources (reduced from 4K to 1920x1080). Can go back up if of interest to anyone. We've since done further optimizations which make this not needed.
• macOS Support (see here)
• Ability to configure what portion of the bounding box is used for averaging depth (Z) and thereby X and Y locations. Configurable via padding_factor, here.
• Add capability to get the die temperatures off of DepthAI VPU. Selectable with meta_d2h stream. See here.
• Fix a segfault on closing streams caused by upgrade to OpenVINO 2020.1 XLink.
• Fix color aliasing effect with some neural models (see here)
• Store calibration, left-to-right, left-to-RGB, etc. board parameters to EEPROM. See PR here and example usage here.
• More rigorous mapping of depth and object detector bounding box (see here). Includes fix w/ balanced center-cropping on grayscale cameras.
• Facial landmark, age, and facial expression networks run standalone.
• Decode CNN structure based on JSON definition.
• Upgrade to OpenVINO to 2020.1
• Capability to pull JPEG from DepthAI API on-demand (see here)
• Capability to pull H.264/H.265-encoded 1080p stream from DepthAI (see here)
• Multi-stage vehicle detection application (available upon request, see here for example of it running)
• Custom training support (see here and here)
• Multi-output-tensor support. Allows running networks which aren't a single output tensor (like MobileNetSSD which is single-tensor).
• Add mirror capability to Calibration flow for more intuitive use.
• Implement timestamp onto all streams to allow host syncing as desired.
• Store calibration to EEPROM with versioning.
• Enable tighter stereo synchronization (now on the order of micro-second-level sync).
• Tie exposure of left and right together for better stereo matching.
• Fix for depth directionality here and here
• Open Source of DepthAI (depthai-api, here)
• Sanitize floating point returns in the C++ depthai-api so Python doesn't have to deal with them. (see )
• Implement USB2 mode (see here)
• Implement basic integration test system (here), which acts as a decent integration example.
• Host watchdog solution, which catches host-side errors (or example errors in coding, uncaught exceptions, etc.) and automatically restarts the script if an exception occurs. See here.
• Fix USB issue w/ Pi 4. See here.
• Allow selectable frame-rate per stream. See PR here and how to use here
• Platform-agnostic DepthAI reset. A watchdog runs in the DepthAI firmware. If the host does not respond within a certain period of time, DepthAI resets itself to wait for the host-side watchdog to trip and get DepthAI running again.
• megaAI (BW1093)
• 3D Object Localization MVP
• Allow user to select individual streams
• Colorized Depth capability
• Reduce frame latency on Raspberry Pi operation to be that of using w/ fast hosts