# QIMP SDK release 1.1

Source: [https://docs.qualcomm.com/doc/80-70015-52/topic/ga1_1_release_information.html](https://docs.qualcomm.com/doc/80-70015-52/topic/ga1_1_release_information.html)

## Release information

Table : Software version

| Software | Version |
| --- | --- |
| Yocto | Kirkstone 4.0.18 |
| Kernel | 6.6.28 |

Table : Release tag version

| Release tag | Version |
| --- | --- |
| Firmware release tag | r1.0\_00039.2 |
| Release tag | qcom-6.6.28-QLI.1.1-Ver.1.1.xml |
| Meta-qcom-extras release tag | r1.0\_00040.0 |
| QIMP SDK release tag | qcom-6.6.28-QLI.1.1-Ver.1.1\_qim-product-sdk-1.1.3 |

| **SoC                                platform** | QCS6490 |
| --- | --- |
| **SoC                                platform** | QCS5430 |
|  |  |

## Reference design kits

Qualcomm^®^ Robotics RB3 Gen 2 Development Kit

## Contents of the release

The contents of the QIMP SDK release include:
- Recipes for building the individual components:
    - Qualcomm IM SDK
    - TensorFlow Lite
    - Qualcomm Neural Processing SDK
    - Qualcomm AI Engine direct SDK
- Sample applications that demonstrate how to use the Qualcomm IM SDK to
                        develop AI edge-based applications.

To get started with the QIMP SDK, see [Qualcomm Intelligent Multimedia Product (QIMP)
                    SDK Quick Start Guide](https://docs.qualcomm.com/bundle/publicresource/topics/80-70015-51/introduction.html).

## New features

The following are the new features introduced in the QIMP SDK release:
- Camera supports:
    - Sample applications
        - `gst-add-streams-as-bundle-example`, which is
                                        a configuration of camera streams as a bundle, thus, saving
                                        on time.
        - `gst-camera-burst-capture-example`, which
                                        allows snapshot capture in the Burst mode.
        - `gst-camera-switch-example`, which allows to
                                        switch between two cameras.
        - `gst-add-remove-streams-runtime`, which
                                        allows the addition and removal of multiple streams
                                        dynamically.
    - Multiple streams with different framerate for each stream
    - USB camera using the v4l2src gstreamer plugin
    - Different video HDR modes: RAW linear SHDR (V2), YUV virtual channel
                                SHDR (V3), and SHDR switch from linear to RAW linear SHDR (V2)
    - Electronic image stabilization (EIS) for preview only or preview +
                                video use case
    - Reprocess to generate multiple streams from a single input stream
                                based on configured region of interest (ROI) coordinates
    - Continuous JPEG snapshots at 30 FPS along with preview
- Video supports:
    - Playback framerate control
- Graphics supports:
    - User configurable polygon privacy mask and inverse privacy mask
                                using the gstqtioverlay plugin
- AI/ML supports:
    - Sample applications
        - `gst-ai-monodepth`, which allows monodepth on
                                        images streamed from a camera.
        - `gst-ai-daisychain-detection-classification`,
                                        which allows cascaded object detection and classification on
                                        images streamed from a camera.
    - Multistage ML pipeline (daisy chaining) for advanced ML use cases;
                                for example, detection followed by classification based on detected
                                objects.
    - Socket-based gstqtisocketsrc and gstqtisocketsink plugins to support
                                efficient/zero copy data flow between pipelines running inside
                                docker and host.
    - Hardware-accelerated gstqtivsplit plugin to support input image crop
                                based on ML detection ROI output. This plugin generates a stream per
                                detected ROI that can be used for multistage ML pipeline or other
                                purposes like storage.
    - Concurrent AI pipelines without impacting the performance
    - Mono-depth estimation ML model (midas-v2) for depth estimation based
                                on a single image

## Sample applications

For a list of sample applications supported in the QIMP SDK and instructions on how
                to run them, see [Sample applications](https://docs.qualcomm.com/bundle/publicresource/topics/80-70015-50/example-applications.html).

## Issues resolved

Fixed speed-based video playback.

## Limitations

The following are the known limitations in the QIMP SDK release:
- When using the qtioverlay plugin with detection models, frame drops may
                        occur, especially with many detections. Use qtivcomposer for detection-based
                        ML use cases.
- A drop of 1-2 FPS may be observed with the three-stream camera use
                        case.

Last Published: Oct 15, 2024

[Previous Topic
QIMP SDK release 1.2](https://docs.qualcomm.com/bundle/publicresource/80-70015-52/topics/qimp-sdk-release-1-2.md) [Next Topic
QIMP SDK release 1.0](https://docs.qualcomm.com/bundle/publicresource/80-70015-52/topics/introduction.md)