st-webcam is a Python package designed to simplify computer vision projects, providing an easy-to-use interface for common computer vision tasks, such as accessing and displaying webcam feeds, applying basic image processing techniques, and integrating with popular libraries like OpenCV and Streamlit. It is perfect for anyone who wants to get started quickly with computer vision applications without dealing with the complexities of managing camera devices and frame handling.
This package contains WebCam, which is a Python class designed to make webcam integration with Streamlit simple and effective. It abstracts away the complexity of accessing and managing webcam feeds, allowing you to focus on building computer vision applications. Whether you're prototyping a computer vision project, experimenting with real-time image processing, or just need a straightforward webcam interface, WebCam offers an easy-to-use solution.
If you find this template useful, please consider giving it a ⭐ on GitHub! It helps others discover this project and lets me know you’re interested.
WebCam Class: Easily integrate and control webcam feeds from various sources.
Simple Webcam Control: Easily start and stop webcam feeds.
Real-time Display: Stream webcam frames in real time within Streamlit apps.
Custom Frame Processing: Apply custom image processing (e.g., filters, effects) to webcam frames before displaying them.
Multi-Webcam Support: Manage multiple webcams by specifying different device indexes.
Session State Management: Leveraging Streamlit’s session state to handle webcam states and resources efficiently.
Lightweight & Beginner-Friendly: Easy-to-understand class-based structure designed for prototyping and learning.
Run the following command to install dependencies.
pip install st-webcam
Import necessary libraries.
import streamlit as st from st_webcam import WebCam # import other required libraries for your project
Run the following command to start your streamlit app.
streamlit run app.py
Where app.py is your Python script that contains the code to display the webcam feed.
Purpose
Arguements
Example
webcam = WebCam(index=1, label="Custom")
Purpose
Starts the webcam feed and initializes the VideoCapture object.
Arguements
Example
webcam.start(index=1)
Purpose
This method releases the webcam resources, clears session state variables, and resets the webcam to a stopped state. If the webcam is not running, it does nothing.
Example
- webcam.stop()
Purpose
Displays the provided frame in the Streamlit interface. Can apply a function before displaying.
Arguements
Example
- webcam.display_frame(frame, frame_func=apply_filter, frame_placeholder=placeholder1)
webcam = WebCam() # webcam object frames = webcam.start() # before use if frames: for frame in frames: webcam.display_frame(frame) webcam.stop() # after use
import cv2 def convert_grayscale(frame): return cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY) webcam = WebCam(index=0) # for webcam at index 0 frames = webcam.start() if frames: for frame in frames: webcam.display_frame(frame, frame_func=convert_grayscale) webcam.stop()
def apply_canny_edge_detection(frame): gray_frame = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY) edges = cv2.Canny(gray_frame, 100, 200) edges_rgb = cv2.cvtColor(edges, cv2.COLOR_GRAY2RGB) return edges_rgb def apply_cartoon_effect(frame): bilateral_filtered_frame = cv2.bilateralFilter(frame, d=9, sigmaColor=75, sigmaSpace=75) gray_frame = cv2.cvtColor(bilateral_filtered_frame, cv2.COLOR_RGB2GRAY) blurred_gray = cv2.medianBlur(gray_frame, 7) cartoon_edges = cv2.adaptiveThreshold(blurred_gray, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, blockSize=9, C=9) cartoon_frame = cv2.bitwise_and(bilateral_filtered_frame, bilateral_filtered_frame, mask=cartoon_edges) return cartoon_frame def apply_sobel_edge_detection(frame): gray_frame = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY) sobel_x = cv2.Sobel(gray_frame, cv2.CV_64F, 1, 0, ksize=3) sobel_y = cv2.Sobel(gray_frame, cv2.CV_64F, 0, 1, ksize=3) sobel_edges = cv2.magnitude(sobel_x, sobel_y) sobel_edges = cv2.convertScaleAbs(sobel_edges) sobel_edges_rgb = cv2.cvtColor(sobel_edges, cv2.COLOR_GRAY2RGB) return sobel_edges_rgb
webcam = WebCam(index=0, label="Cartoon") frames = webcam.start() placeholder1 = st.empty() placeholder2 = st.empty() if frames: for frame in frames: webcam.display_frame(frame, apply_canny_edge_detection) webcam.display_frame(frame, apply_cartoon_effect, placeholder1) webcam.display_frame(frame, apply_sobel_edge_detection, placeholder2) webcam.stop()
Feel free to fork the project, contribute, or create an issue for any bugs or new features you'd like to see. If you're interested in collaborating, please follow the standard GitHub contribution workflow: fork, clone, create a branch, and submit a pull request.
st-webcam is licensed under the MIT License. See the License file for more details.
