uwdb / visualroad Goto Github PK
View Code? Open in Web Editor NEWVisual Road: A Video Data Management Benchmark
Home Page: http://visualroad.uwdb.io
License: MIT License
Visual Road: A Video Data Management Benchmark
Home Page: http://visualroad.uwdb.io
License: MIT License
Line 74 in dd6c20d
Line 75 in dd6c20d
change L74 and L75 to:
RUN pip2 install --user pyyaml opencv-python==4.0.1.24 psutil & \ pip3 install --user pyyaml opencv-python==4.0.1.24 psutil
Due to the following error in the latest package opevcv-python
https://pypi.org/project/opencv-python/
Parameter | Value |
---|---|
Name | 1K-Short-1 |
Benchmark version | 1.0 |
Dataset version | 1.0 |
Seed | 0x00181CD |
Scale | 1 |
Resolution | 960 x 540 |
Duration | 15 |
Download link | http://homes.cs.washington.edu/~bhaynes/datasets/short-1k.tar.gz |
Parameter | Value |
---|---|
Name | 4k-short-1 |
Benchmark version | 1.0 |
Dataset version | 1.0 |
Seed | 0x00181CD |
Scale | 1 |
Resolution | 960 x 540 |
Duration | 15 |
Download link | link |
Hello, I just found out this great project, but I noticed that it doesn't specify which version of the ue4-docker Python package it is using, neither what configuration was used to generate the image adamrehn/ue4-engine:4.22.0
that you are using in your Dockerfile.
I would guest that this is the ue4-engine only image (https://adamrehn.com/docs/ue4-docker/building-images/available-container-images#ue4-engine) build with CUDA instead of OpenGL (I noticed that the nvidia runtime is specified in the dockerfile), is this correct?
And if using CUDA, what was the version used?
Sorry for all the questions, and thank you for this great work.
Hi visualroad team,
As described on your project website, "Visual Road's video data is automatically generated with a high degree of realism, and annotated using a modern simulation and visualization engine" . I downloaded the videos/Pregenerated Datasets here but I can not find the annotations files, could you please provide me those annotations files?
Thank you very much!
On some configurations containing very large numbers of vehicles, a "traffic deadlock" can occur where no vehicle makes progress. When no camera observes the stationary vehicles, this may appear as if the vehicles have vanished (especially when no pedestrians are present).
Reported by Chris Liu @ HKU.
Parameter | Value |
---|---|
Name | 1K-Short-2 |
Benchmark version | 1.0 |
Dataset version | 1.0 |
Seed | 0x75BCD15 |
Scale | 2 |
Resolution | 960 x 540 |
Duration | 15 |
Download link | http://homes.cs.washington.edu/~bhaynes/datasets/short-1k-2.tar.gz |
Comments:
Parameter | Value |
---|---|
Name | 1K-Long-4 |
Benchmark version | 1.0 |
Dataset version | 1.0 |
Seed | 0x75BCD15 |
Scale | 4 |
Resolution | 1K (960x540) |
Duration | 60 |
Download link | link |
Hi, As you mentioned in your paper, that we could get the ground-truth/bounding box from the video, so how can I generate it?
Please show me the command or some needed modification/configuration to generate it if possible! Thank you very much!
Hi there, I've managed to generate a simple 3 seconds simulation, and the driver output, to tests things out. But when got to the Verifying Query Results
I've noticed that there's no verifier.py file.
Maybe it was not committed to the repository?
Thanks
I am currently trying to test queries in offline settings.
While looking into verifier.py I realized few functions(q2c, q3, q6a, q6b/ q7~10) haven't been implemented yet.
Would it be possible to know any future plans for those verifier functions, if any?
Best regards
As mentioned in the paper, I wanted to find out the implementations for the vdbms queries' benchmarking done for systems such as Scanner, LightDB and NoScope.
Hello, I am attracted by the vehicle tracking query mentioned in the paper visualroad. I have several questions and hope to be answered.
I want to know all the vehicle tracks in the dataset, when does a vehicle pass by a camera, eg vehicle_id1 passes by camera_id3 during time 150s to 160s, vehicle_id1 passes by camera_id8 during time 230s to 225s. During the time 160s to 230s, vehicle_id1 is on the road so that no camera records it.
So I have the following questions:
I would feel honored to get a reply.
Hi,
Thanks for the great work!
Recently I've been working on processing specific queries on videos and I would like to use your generator as a benchmark. However, when I try to generator the dataset with scale=2
, it stopped during generation and gave me an error.
Could you please help me out? Am I missing something? Many thanks!
The command I am using is sudo docker-compose run generator --scale 2 dataset-test-2
The output is:
4.22.0-0+++UE4+Release-4.22 517 0
Disabling core dumps.
INFO:root:Map: Town03, Weather: WeatherParameters(cloudyness=80.000000, precipitation=60.000000, precipitation_deposits=100.000000, wind_intensity=1.000000, sun_azimuth_angle=0.000000, sun_altitude_angle=15.000000), Vehicles: 200, Walkers: 250
INFO:root:Tile 1 of 2: Rendered 0 frames; 900 remaining (0.0 FPS)
INFO:root:Tile 1 of 2: Rendered 0 frames; 900 remaining (0.0 FPS)
INFO:root:Tile 1 of 2: Rendered 0 frames; 900 remaining (0.0 FPS)
INFO:root:Tile 1 of 2: Rendered 0 frames; 900 remaining (0.0 FPS)
INFO:root:Tile 1 of 2: Rendered 0 frames; 900 remaining (0.0 FPS)
INFO:root:Tile 1 of 2: Rendered 0 frames; 900 remaining (0.0 FPS)
INFO:root:Tile 1 of 2: Rendered 0 frames; 900 remaining (0.0 FPS)
INFO:root:Tile 1 of 2: Rendered 0 frames; 900 remaining (0.0 FPS)
INFO:root:Tile 1 of 2: Rendered 0 frames; 900 remaining (0.0 FPS)
INFO:root:Tile 1 of 2: Rendered 0 frames; 900 remaining (0.0 FPS)
INFO:root:Tile 1 of 2: Rendered 8 frames; 892 remaining (0.2 FPS)
INFO:root:Tile 1 of 2: Rendered 18 frames; 882 remaining (0.3 FPS)
INFO:root:Tile 1 of 2: Rendered 28 frames; 872 remaining (0.5 FPS)
INFO:root:Tile 1 of 2: Rendered 38 frames; 862 remaining (0.6 FPS)
INFO:root:Tile 1 of 2: Rendered 48 frames; 852 remaining (0.7 FPS)
INFO:root:Tile 1 of 2: Rendered 58 frames; 842 remaining (0.8 FPS)
INFO:root:Tile 1 of 2: Rendered 68 frames; 832 remaining (0.9 FPS)
INFO:root:Tile 1 of 2: Rendered 78 frames; 822 remaining (1.0 FPS)
INFO:root:Tile 1 of 2: Rendered 88 frames; 812 remaining (1.0 FPS)
INFO:root:Tile 1 of 2: Rendered 98 frames; 802 remaining (1.1 FPS)
INFO:root:Tile 1 of 2: Rendered 108 frames; 792 remaining (1.1 FPS)
INFO:root:Tile 1 of 2: Rendered 118 frames; 782 remaining (1.2 FPS)
INFO:root:Tile 1 of 2: Rendered 128 frames; 772 remaining (1.2 FPS)
INFO:root:Tile 1 of 2: Rendered 138 frames; 762 remaining (1.3 FPS)
INFO:root:Tile 1 of 2: Rendered 148 frames; 752 remaining (1.3 FPS)
INFO:root:Tile 1 of 2: Rendered 158 frames; 742 remaining (1.4 FPS)
INFO:root:Tile 1 of 2: Rendered 168 frames; 732 remaining (1.4 FPS)
INFO:root:Tile 1 of 2: Rendered 178 frames; 722 remaining (1.4 FPS)
INFO:root:Tile 1 of 2: Rendered 188 frames; 712 remaining (1.4 FPS)
INFO:root:Tile 1 of 2: Rendered 198 frames; 702 remaining (1.5 FPS)
INFO:root:Tile 1 of 2: Rendered 208 frames; 692 remaining (1.5 FPS)
INFO:root:Tile 1 of 2: Rendered 218 frames; 682 remaining (1.5 FPS)
INFO:root:Tile 1 of 2: Rendered 228 frames; 672 remaining (1.5 FPS)
INFO:root:Tile 1 of 2: Rendered 238 frames; 662 remaining (1.6 FPS)
INFO:root:Tile 1 of 2: Rendered 248 frames; 652 remaining (1.6 FPS)
INFO:root:Tile 1 of 2: Rendered 258 frames; 642 remaining (1.6 FPS)
INFO:root:Tile 1 of 2: Rendered 268 frames; 632 remaining (1.6 FPS)
INFO:root:Tile 1 of 2: Rendered 278 frames; 622 remaining (1.6 FPS)
INFO:root:Tile 1 of 2: Rendered 288 frames; 612 remaining (1.7 FPS)
INFO:root:Tile 1 of 2: Rendered 298 frames; 602 remaining (1.7 FPS)
INFO:root:Tile 1 of 2: Rendered 308 frames; 592 remaining (1.7 FPS)
INFO:root:Tile 1 of 2: Rendered 318 frames; 582 remaining (1.7 FPS)
INFO:root:Tile 1 of 2: Rendered 328 frames; 572 remaining (1.7 FPS)
INFO:root:Tile 1 of 2: Rendered 338 frames; 562 remaining (1.7 FPS)
INFO:root:Tile 1 of 2: Rendered 348 frames; 552 remaining (1.7 FPS)
INFO:root:Tile 1 of 2: Rendered 358 frames; 542 remaining (1.7 FPS)
INFO:root:Tile 1 of 2: Rendered 368 frames; 532 remaining (1.8 FPS)
INFO:root:Tile 1 of 2: Rendered 378 frames; 522 remaining (1.8 FPS)
INFO:root:Tile 1 of 2: Rendered 388 frames; 512 remaining (1.8 FPS)
INFO:root:Tile 1 of 2: Rendered 398 frames; 502 remaining (1.8 FPS)
INFO:root:Tile 1 of 2: Rendered 408 frames; 492 remaining (1.8 FPS)
INFO:root:Tile 1 of 2: Rendered 418 frames; 482 remaining (1.8 FPS)
INFO:root:Tile 1 of 2: Rendered 428 frames; 472 remaining (1.8 FPS)
INFO:root:Tile 1 of 2: Rendered 438 frames; 462 remaining (1.8 FPS)
INFO:root:Tile 1 of 2: Rendered 448 frames; 452 remaining (1.8 FPS)
INFO:root:Tile 1 of 2: Rendered 458 frames; 442 remaining (1.8 FPS)
INFO:root:Tile 1 of 2: Rendered 468 frames; 432 remaining (1.8 FPS)
INFO:root:Tile 1 of 2: Rendered 478 frames; 422 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 488 frames; 412 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 498 frames; 402 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 508 frames; 392 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 518 frames; 382 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 528 frames; 372 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 538 frames; 362 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 548 frames; 352 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 558 frames; 342 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 568 frames; 332 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 578 frames; 322 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 588 frames; 312 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 598 frames; 302 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 608 frames; 292 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 618 frames; 282 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 628 frames; 272 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 638 frames; 262 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 648 frames; 252 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 658 frames; 242 remaining (1.9 FPS)
INFO:root:Tile 1 of 2: Rendered 668 frames; 232 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 678 frames; 222 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 688 frames; 212 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 698 frames; 202 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 708 frames; 192 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 718 frames; 182 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 728 frames; 172 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 738 frames; 162 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 748 frames; 152 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 758 frames; 142 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 768 frames; 132 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 778 frames; 122 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 788 frames; 112 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 798 frames; 102 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 808 frames; 92 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 818 frames; 82 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 828 frames; 72 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 838 frames; 62 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 848 frames; 52 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 858 frames; 42 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 868 frames; 32 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 878 frames; 22 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 888 frames; 12 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 898 frames; 2 remaining (2.0 FPS)
INFO:root:Tile 1 of 2: Rendered 908 frames; -8 remaining (2.0 FPS)
INFO:root:Destroying actors
INFO:root:Generation complete for tile 0
INFO:root:Map: Town07, Weather: WeatherParameters(cloudyness=80.000000, precipitation=30.000000, precipitation_deposits=50.000000, wind_intensity=0.400000, sun_azimuth_angle=0.000000, sun_altitude_angle=75.000000), Vehicles: 100, Walkers: 400
Traceback (most recent call last):
File "/app/generator.py", line 409, in <module>
generate(args.path, tile_pool, args.scale, (args.width, args.height), args.duration, args.fov, args.seed)
File "/app/generator.py", line 345, in generate
generate_tile(client, path, id, used_tiles[-1], scale, resolution, duration, panorama_fov)
File "/app/generator.py", line 232, in generate_tile
client.load_world(tile.map)
RuntimeError: map not found
FUnixPlatformMisc::RequestExitWithStatus
FUnixPlatformMisc::RequestExit
The generated configuration file is
duration: 30
hostname: localhost
name: dataset-test-2
panorama_fov: null
port: 2000
resolution:
height: 540
width: 960
scale: 2
seed: 0
tiles:
- cameras:
- type: traffic
videos:
- /app/dataset-test-2/traffic-000.mp4
- type: traffic
videos:
- /app/dataset-test-2/traffic-001.mp4
- type: traffic
videos:
- /app/dataset-test-2/traffic-002.mp4
- type: traffic
videos:
- /app/dataset-test-2/traffic-003.mp4
- type: panoramic
videos:
- /app/dataset-test-2/panoramic-000-000.mp4
- /app/dataset-test-2/panoramic-000-001.mp4
- /app/dataset-test-2/panoramic-000-002.mp4
- /app/dataset-test-2/panoramic-000-003.mp4
id: 0
map: Town03
pedestrians: 250
vehicles: 200
weather: WeatherParameters(cloudyness=80.000000, precipitation=60.000000, precipitation_deposits=100.000000,
wind_intensity=1.000000, sun_azimuth_angle=0.000000, sun_altitude_angle=15.000000)
- cameras:
- type: traffic
videos:
- /app/dataset-test-2/traffic-004.mp4
- type: traffic
videos:
- /app/dataset-test-2/traffic-005.mp4
- type: traffic
videos:
- /app/dataset-test-2/traffic-006.mp4
- type: traffic
videos:
- /app/dataset-test-2/traffic-007.mp4
- type: panoramic
videos:
- /app/dataset-test-2/panoramic-001-000.mp4
- /app/dataset-test-2/panoramic-001-001.mp4
- /app/dataset-test-2/panoramic-001-002.mp4
- /app/dataset-test-2/panoramic-001-003.mp4
id: 1
map: Town07
pedestrians: 400
vehicles: 100
weather: WeatherParameters(cloudyness=80.000000, precipitation=30.000000, precipitation_deposits=50.000000,
wind_intensity=0.400000, sun_azimuth_angle=0.000000, sun_altitude_angle=75.000000)
timeout: 30
version: 1.0
Hello, first of all thank you for your amazing work.
I have been trying to generate some visualroad data following the steps listed in github.
However, while running the 'docker-compose run generator --scale 1 my-dataset' command I am facing some issues.
FULL OUTPUT OF ue4-docker info :
ue4-docker version: 0.0.62 (latest available version is 0.0.62)
Operating system: Linux (Ubuntu 18.04.4 LTS, 5.3.0-46-generic)
Docker daemon version: 19.03.8
NVIDIA Docker supported: No
Maximum image size: No limit detected
Available disk space: 99.22 GiB
Total system memory: 15.59 GiB physical, 2 GiB virtual
Number of processors: 4 physical, 8 logical
FULL OUTPUT OF docker-compose run generator --scale 1 my-dataset
error: XDG_RUNTIME_DIR not set in the environment.
Signal 11 caught.
Malloc Size=65538 LargeMemoryPoolOffset=65554
Malloc Size=65535 LargeMemoryPoolOffset=131119
Malloc Size=48848 LargeMemoryPoolOffset=179984
4.22.0-0+++UE4+Release-4.22 517 0
Disabling core dumps.
CommonUnixCrashHandler: Signal=11
Engine crash handling finished; re-raising signal 11 for the default handler. Good bye.
INFO:root:Map: Town03, Weather: WeatherParameters(cloudyness=80.000000, precipitation=60.000000, precipitation_deposits=100.000000, wind_intensity=1.000000, sun_azimuth_angle=0.000000, sun_altitude_angle=15.000000), Vehicles: 200, Walkers: 250
Traceback (most recent call last):
File "./generator.py", line 425, in <module>
generate(args.path, tile_pool, args.scale, (args.width, args.height), args.duration, args.fov, args.seed, args.vehicles, args.pedestrians)
File "./generator.py", line 349, in generate
generate_tile(client, path, id, used_tiles[-1], scale, resolution, duration, panorama_fov)
File "./generator.py", line 232, in generate_tile
client.load_world(tile.map)
RuntimeError: time-out of 30000ms while waiting for the simulator, make sure the simulator is ready and connected to localhost:2000
Would there be any suggestions on how to fix this issue, please?
Best regards.
Parameter | Value |
---|---|
Name | 1K-Short-2 |
Benchmark version | 1.0 |
Dataset version | 1.0 |
Seed | 0x75BCD15 |
Scale | 2 |
Resolution | 1920 x 1080 |
Duration | 15 |
Download link | http://homes.cs.washington.edu/~bhaynes/datasets/short-2k-2.tar.gz |
When I visit the home web https://db.cs.washington.edu/projects/visualroad/ trying to download the pregenerated datasets, I find that only the dataset 1K-Long-4
is available.
The other four datasets located on http://homes.cs.washington.edu/~bhaynes/datasets are unable to download.
Would you please share a google drive link for those datasets? Thank you so much.
I have been trying to setup visualroad for synthetic dataset generation.
I completed all the steps till the part where I run the generator service, which fails as:
anip@blade:~/codes/visualroad$ docker-compose run generator --scale 1 my-dataset
Pulling generator (visualroad/core:latest)...
ERROR: pull access denied for visualroad/core, repository does not exist or may require 'docker login': denied: requested access to the resource is denied
Can anyone help me with this? Is it due to some access issue.
EDIT 1):
Looks like the issue was because the docker build docker build -t visualroad/core .
did not execute successfully. It fails in one of the steps like:
[2/2] cd /home/ue4/carla/Build/llvm-7.0-build/projects/libcxx/lib && /usr/bin/cmake -DCMAKE_INSTALL_COMPONENT=cxx -P /home/ue4/carla/Build/llvm-7.0-build/projects/libcxx/cmake_install.cmake
-- Install configuration: "RelWithDebInfo"
-- Installing: /home/ue4/carla/Build/llvm-7.0-install/lib/libc++.so.1.0
-- Installing: /home/ue4/carla/Build/llvm-7.0-install/lib/libc++.so.1
-- Installing: /home/ue4/carla/Build/llvm-7.0-install/lib/libc++.so
-- Installing: /home/ue4/carla/Build/llvm-7.0-install/lib/libc++.a
[1/2] Linking CXX static library lib/libc++abi.a
[2/2] cd /home/ue4/carla/Build/llvm-7.0-build/projects/libcxxabi/src && /usr/bin/cmake -DCMAKE_INSTALL_COMPONENT=cxxabi -P /home/ue4/carla/Build/llvm-7.0-build/projects/libcxxabi/cmake_install.cmake
-- Install configuration: "RelWithDebInfo"
-- Installing: /home/ue4/carla/Build/llvm-7.0-install/lib/libc++abi.so.1.0
-- Installing: /home/ue4/carla/Build/llvm-7.0-install/lib/libc++abi.so.1
-- Installing: /home/ue4/carla/Build/llvm-7.0-install/lib/libc++abi.so
-- Installing: /home/ue4/carla/Build/llvm-7.0-install/lib/libc++abi.a
Setup.sh: Retrieving boost.
--2022-04-01 14:16:52-- https://dl.bintray.com/boostorg/release/1.69.0/source/boost_1_69_0.tar.gz
Resolving dl.bintray.com (dl.bintray.com)... 35.161.19.108, 35.83.99.37
Connecting to dl.bintray.com (dl.bintray.com)|35.161.19.108|:443... connected.
HTTP request sent, awaiting response... 502 Bad Gateway
2022-04-01 14:16:52 ERROR 502: Bad Gateway.
Util/BuildTools/Linux.mk:110: recipe for target 'setup' failed
make: *** [setup] Error 8
The command '/bin/sh -c git clone https://github.com/carla-simulator/carla $CARLA_PATH && cd $CARLA_PATH && git checkout $CARLA_VERSION && ./Update.sh && make package' returned a non-zero code: 2
It looks like the build is trying to access an outdated bintray
link. Any help would be appreciated. Thanks.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.