Skip to main content

🌈 Vancouver Watching with AI.

Project description

🌈 Vancouver Watching (vanwatch)

vanwatch 🌈 discovers and ingests images from traffic cameras in an area and then runs YOLO 🚀, OpenAI Vision, and other vision algo to extract information about urban activity at scale.

pip install vancouver-watching
 > vanwatch help
vanwatch conda create [validate,~recreate]
 . create conda environment.
vanwatch conda validate
 . validate conda environment.
vanwatch discover \
	[area=<area>,~upload] \
	[-|<object-name>] \
	[<args>]
 . discover area -> <object-name>.
vanwatch ingest \
	area=<area>,count=<count>,dryrun,gif,model=<model-id>,~process,publish,~upload \
	-|<object-name> \
	[<args>]
 . ingest <area> -> <object-name>.
vanwatch list [area=<area>,discovery|ingest,published] \
	[--count <count>] \
	[--delim space] \
	[--log 0] \
	[--offset <offset>]
 . list objects from area.
2 area(s): iran,vancouver
vanwatch list areas
 . list areas.
vanwatch vision "prompt" \
	[area=<area>,offset=<1>,auto|low|high,dryrun,~upload] \
	Davie,Bute \
	[--verbose 1]
 . openai_commands vision: prompt @ <area>/intersection.
vanwatch process \
	count=<count>,~download,gif,model=<model-id>,publish,~upload \
	.|<object-name> \
	[--detect_objects 0] \
	[--do_dryrun 1] \
	[--overwrite 1] \
	[--verbose 1]
 . process <object-name>.
vanwatch pylint
 . pylint vancouver_watching.
vanwatch update|update_cache \
	area=<vancouver>,overwrite,process,~publish,refresh,~upload \
	[--verbose 1]
 . update QGIS cache.
vancouver_watching test \
	[dryrun,~ingest,~list,~process,upload]
 . test vancouver_watching.

last build 🔗 image

discover and Ingest an Area

image

to see the list of areas supported by vanwatch type in,

vanwatch list areas

to discover the available cameras in an area type in,

vanwatch discover area=vancouver

you have generated a geojson of traffic images in the City of Vancouver. Now, you can ingest the traffic images from this area and detect people and cars in them,

vanwatch ingest area=vancouver,count=2,publish

image

model: https://hub.ultralytics.com/models/R6nMlK6kQjSsQ76MPqQM?tab=preview

image

image

image

dataset: vanwatch-cache-2024-02-28-21-04-19-26236.tar.gz (details).


PyPI version

to use on AWS SageMaker replace <plugin-name> with vanwatch and follow these instructions.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

vancouver_watching-3.394.1.tar.gz (18.1 kB view details)

Uploaded Source

Built Distribution

vancouver_watching-3.394.1-py3-none-any.whl (25.8 kB view details)

Uploaded Python 3

File details

Details for the file vancouver_watching-3.394.1.tar.gz.

File metadata

  • Download URL: vancouver_watching-3.394.1.tar.gz
  • Upload date:
  • Size: 18.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.9.19

File hashes

Hashes for vancouver_watching-3.394.1.tar.gz
Algorithm Hash digest
SHA256 f6af03bae94ba46597a15546869dce2632a6f7aedc8115d082afa7864a62e544
MD5 a7e2feb605fa850c3d27e0703a13cb89
BLAKE2b-256 33089f3f54bd4ee04b3a7e2d71dee65d453b1cf0b16500af88ce5123297ba00c

See more details on using hashes here.

File details

Details for the file vancouver_watching-3.394.1-py3-none-any.whl.

File metadata

File hashes

Hashes for vancouver_watching-3.394.1-py3-none-any.whl
Algorithm Hash digest
SHA256 f8b92147efedca55827b3d0e305a9d29e5c1e72edf04ef7ced98828089d2aff8
MD5 88417954e5ec771c27ed6cf76f3c2131
BLAKE2b-256 655b63caa60d2e071be00928df774fb2ff33213b308135c81f49177cd52f708c

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page