Vision Apps Release Notes
Version: 10.01.00
Contents
- Introduction
- Licensing
- Getting Started
- Documentation
- What's New
- Upgrade and Compatibility Information
- Device Support
- Validation Information
- Fixed Issues
- Open Defects
- Technical Support
- Package Versioning
Introduction
The Vision Apps package consists of ADAS, vision, deep learning, perception demos/applications for Jacinto 7 platform
To run the demos in vision apps, the companion Processor SDK Linux for Jacinto 7 also needs to be downloaded separately.
Licensing
Refer to Processor SDK RTOS manifest at top level [HTML]
Getting Started
The Vision Apps User Guide [HTML] provides the documentation and references necessary to begin development on TI's platforms using Vision Apps.
Documentation
Refer to following documentation for further details:
| Vision Apps User Guide | Build instructions, API Guide | [HTML] |
| Surround View Manual Calibration Tool | Instructions for Generating Calibration Binaries | [PDF] |
| Release Notes Archive | Previous release notes | [FOLDER] |
What's New
New features in this release are highlighted in bold- OpenVX based demos for ADAS, Vision, deep learning applications
- Free RTOS and Safe RTOS on C7x, C6x on R5F and Linux or QNX on A72
- Integrates all major PSDK RTOS and PSDK Linux SW components like TIDL, MMALIB, PDK, OpenVX, ETHFW, OpenGL, C6x algorithms, imaging/sensors, IPC, Linux, TI-RTOS, SPL/uboot.
- Integrates all major HW components like CSI2 camera, eDP/HDMI display, UART, I2C, ethernet, SD card
- Deep Learning demos
- Image Classification demo
- File based
- Camera based
- Standalone Semantic Segmentation application
- File based
- Camera based
- Standalone Object Detection application
- File based
- Camera based
- Auto valet parking (AVP) demo
- semantic segmentation, Parking spot detect and Vehicle detect algorithms
- 1 channel, 3 algo mode
- 3 channel, 3 algo mode (combined network for parking spot and vehicle detect)
- 3 channel, 5 algo mode - combined network for parking spot and vehicle detect + Three task network for Semantic Segmentation, Motion Segmentation and Depth Estimation (AVP3 demo)
- DKAZE-based Visual Localization application
- Debug support for the ability to debug intermediate layer information from TIDL node
- Surround View demos
- 3D surround view using GPU on 4x 2MP 30fps live camera input
- 3D surround view calibration application
- 3D Surround view + AVP2 demo running simultaneously
- 3D surround view using GPU on 4x 2MP 30fps live camera input
- 3 channel, 3 algo mode (combined network for parking spot and vehicle detect)
- ADAS/Vision demos
- Dense optical flow (DOF) demo
- Stereo disparity engine demo
- Single and multi camera CSI2RX + VISS + LDC + MSC + Display demo
- Multi channel codec demo utilizing CSI2RX + VISS + LDC + MSC + Display with Linux and QNX running on A72
- C7x algorithm offload demo (with DMA acceleration)
- CV Demos
- Structure from motion
- OpenVX target nodes
- image pre-processing (YUV 2 RGB) nodes for DL demos
- image post-processing and visualization nodes for deep learning, DOF, Stereo demos
- lidar, radar, camera SFM nodes for PTK demos
- OpenGL nodes for surround view
- Multi-threading support in MSC mosaic node
- Block based processing example kernel on C66x using UDMA/DRU
- Profiling, logging and other utility APIs
- General stability, performance, documentation improvements
- All included MCU tasks are split between MCU2_0 (ETHFW, VPAC HWA) and MCU2_1 (DMPAC HWA), leaving MCU3_0 and MCU3_1 free on main island
- VISS Node Optimizations and performance related bug fixes
- [Optimization] Moved VISS configuration memory to cache write-through region to save 125us of configuration time per camera frame (eliminating cache write-back maintenance operation time)
- [Perf Bug Fix] Removed unnecessary cache invalidate maintenance operation on the VISS configuration memory to save 125us of configuration time per camera frame
- [Perf Bug Fix] Removed redundant cpu copies of DCC tables per camera frame (EdgeEnhancement, gamma tables, raw histogram lut)
Upgrade and Compatibility Information
No upgrade and compatibility changes since previous release.
Device Support
Refer Processor SDK RTOS release notes [HTML]
Validation Information
The tools used to validate this package are listed in top level user guide [HTML]
Fixed Issues
| ID | Summary |
|---|---|
| ADASVISION-6266 | tivxWriteImageNode incorrect save for YUV422 format |
| ADASVISION-6209 | The vision apps multicam codec application is not working |
| ADASVISION-6136 | libC7120-host-emulation not used for x86 emulation |
| ADASVISION-6108 | Fix False Positive in QNX SPL with MCU1_0 Enabled |
| ADASVISION-5912 | Fix False Results on Vision Apps Init Test Scriplet |
| ADASVISION-6408 | Test mode in single cam and multi-cam vision_apps demo is broken |
| ADASVISION-6404 | Multi-cam non-interactive mode fails to start when channel mask does not include port 0 |
| ADASVISION-6397 | [Addr-Map]: reduced size of macro in comparison statement in appTarget2SharedConversion API |
| ADASVISION-6350 | PSDK RTOS should use the toolchain from PSDK Linux installer |
| ADASVISION-6014 | [CSIRX]: CSIRX short frame error with erroneous frame is observed in high BW usecase |
Open Defects
| ID | Summary |
|---|---|
| ADASVISION-6453 | SRV + AVP demo not working |
| ADASVISION-6451 | mailbox queue overflowing while running vx_app_arm_ipc.out |
| ADASVISION-6436 | Extend multicam support to any channel mask instead of contiguous |
| ADASVISION-6420 | Perf_stats does not caculate CPU load for A72 on QNX |
| ADASVISION-6387 | No option to set vpParams->scanFmt in the dss utility from vision_apps |
| ADASVISION-5497 | J721e: Ethernet driver crash while running iperf3 after about 20 minutes |
| ADASVISION-5918 | Codec app compiler warnings with GCC 11.3 |
| ADASVISION-5673 | Permission issue can occur when running the vision apps script incorrectly |
| ADASVISION-5669 | SCIserver should use UART logs instead of appLogPrintf |
| ADASVISION-5555 | SDE disparity artifacts on EVM |
| ADASVISION-5393 | AVP4 freezes in first second of running |
| ADASVISION-5100 | Peak DDR BW numbers seem too high for some demos |
| ADASVISION-5034 | [Doc]: Details about 8-channels/camera usecase |
| ADASVISION-4923 | Single cam error on ports other than port 1 after running multi cam |
| ADASVISION-4196 | SRV crash occasionally observed |
| ADASVISION-4080 | QNX: SRV File I/O failing with specific car model |
| ADASVISION-6199 | Race Conditions in c7x-mma-tidl and tiadalg builds |
Technical Support
For technical support and additional assistance, contact local TI Field Application Engineer
Package Versioning
Each package version is composed of 4 period-delimited numbers - represented here by the letters M, m, p and b [M.m.p.b]. The table below provides a descriptive reference regarding package version numbering.
| Digit | Meaning | Description |
|---|---|---|
| 1 (M=Major) | Major revision | Incremented when the new version is substantially different from the previous For example, a new module added or an existing module's algorithm significantly altered. |
| 2 (m=minor) | Minor revision | Incremented when the new version has changed but not in a major way. For example, some minor changes in the API or feature set. |
| 3 (p=patch) | Patch number | Incremented for all other source code changes. This include any packaging support code. |
| 4 (b=build) | Build number | Incremented for each release delivery to CM. Reset for any change to M, m or p |
Copyright 2020, Texas Instruments Incorporated