Smart Crowd Counter

A Streamlit app that counts people in conference session photos using Snowflake Cortex AISQL. It detects raised hands too, which is handy for tracking badge giveaways or audience engagement.

What it does

• Counts attendees in uploaded photos
• Detects raised hands
• Shows conversion rates (hands up vs total people)
• Displays images with interactive analytics

Tech Stack

• Snowflake Cortex AISQL for image analysis
• Streamlit in Snowflake for the app
• Snowflake CLI for setup automation
• Python, Altair for charts

Prerequisites

• Snowflake account with Cortex AISQL enabled (sign up here if you don't have one)
• Snowflake CLI installed and configured
• Task (optional, for automation)

Setup

1. Configure Environment

Copy the example environment file and update with your values:

cp .env.example .env

Edit .env with your Snowflake connection details:

# Snowflake Connection
SNOWFLAKE_ACCOUNT=your_account
SNOWFLAKE_USER=your_username
SNOWFLAKE_ROLE=your_role
SNOWFLAKE_WAREHOUSE=your_warehouse

# Application Configuration
SNOWFLAKE_DATABASE=CROWD_COUNTER_DB
SNOWFLAKE_SCHEMA=CONFERENCES
SNOWFLAKE_STAGE=SNAPS

# AI Model
AI_MODEL=claude-4-sonnet

2. Run Setup

Choose one of the following methods:

Option A: Using Task (recommended)

# Verify Snowflake CLI connection
task snow:check

# Preview rendered SQL (dry run)
task snow:setup:dry-run

# Create all Snowflake objects
task snow:setup

Option B: Using Snowflake CLI

The setup script uses Jinja2 templating for variable substitution:

snow sql -f setup.sql --templating=ALL \
  -D database=CROWD_COUNTER_DB \
  -D schema=CONFERENCES \
  -D stage=SNAPS \
  -D ai_model=claude-4-sonnet

Option C: Using Snowflake Workspaces (No CLI required)

If you prefer working directly in Snowflake's UI without installing any CLI tools:

1. Open Snowsight and navigate to Worksheets
2. Create a new SQL Worksheet in your Workspace
3. Copy the contents of setup.sql into the worksheet
4. Replace the Jinja2 variables with your values:
   • {{ db }} → CROWD_COUNTER_DB (or your database name)
   • {{ sch }} → CONFERENCES (or your schema name)
   • {{ stg }} → SNAPS (or your stage name)
   • {{ model }} → claude-4-sonnet (or your preferred model)
5. Run all statements (Ctrl/Cmd + Shift + Enter)

Tip: Use Find and Replace (Ctrl/Cmd + H) to quickly substitute all variables.

Option D: Using EXECUTE IMMEDIATE FROM

Upload setup.sql to a stage, then run:

EXECUTE IMMEDIATE FROM @my_stage/setup.sql
  USING (database => 'CROWD_COUNTER_DB',
         schema => 'CONFERENCES',
         stage => 'SNAPS',
         ai_model => 'claude-4-sonnet');

3. Deploy the App

Upload app.py to your Snowflake environment and create a Streamlit app using Snowflake's native Streamlit support.

Usage

With the app:

1. Select your database and schema from the dropdowns
2. Upload conference photos (JPG, PNG, JPEG)
3. Wait for the analysis to complete
4. Click on any row to see the image and detailed charts

From command line:

# Upload images
task snow:upload -- /path/to/photos/*.jpg

# List uploaded files
task snow:list

# Query results
task snow:query

Available Tasks

Run task to see all available commands:

Task	Description
task snow:check	Verify Snowflake CLI is installed and configured
task snow:setup	Create all Snowflake objects
task snow:setup:dry-run	Preview rendered SQL without executing
task snow:upload -- FILES	Upload images to stage
task snow:list	List files in stage
task snow:query	Query the crowd counter view
task snow:refresh	Refresh stage directory
task snow:clean	Remove files from stage
task snow:teardown	Drop all created objects (DESTRUCTIVE)

Configuration

All configuration is managed through environment variables in .env:

Variable	Default	Description
SNOWFLAKE_DATABASE	CROWD_COUNTER_DB	Target database
SNOWFLAKE_SCHEMA	CONFERENCES	Target schema
SNOWFLAKE_STAGE	SNAPS	Stage for images
AI_MODEL	claude-4-sonnet	Cortex AI model

Project Structure

smart-crowd-counter/
├── app.py              # Streamlit application
├── setup.sql           # Snowflake object creation
├── Taskfile.yml        # Task automation
├── environment.yml     # Conda dependencies
├── .env.example        # Environment template
├── LICENSE             # Apache 2.0
├── NOTICE              # Attribution notices
└── CITATION.cff        # Citation metadata

Acknowledgments

• Snowflake for Cortex AISQL
• Streamlit for making web apps easy

Using This Project

This project is open source under Apache 2.0. You're welcome to use, modify, and share it!

If you use this project, please:

1. Keep the copyright notices and NOTICE file intact
2. Credit the original author in your README or documentation:

   Based on Smart Crowd Counter by Kamesh Sampath

3. If presenting at events or demos, mention the original creator
4. Consider starring the repo if it helped you

For academic or formal citations, click the "Cite this repository" button on GitHub or see CITATION.cff.

License

Copyright 2026 Kamesh Sampath

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

See the LICENSE and NOTICE files for details.