add pose estimation example

Author: ks6088ts

apps/99_streamlit_examples/README.md

@@ -73,6 +73,10 @@ Access to http://localhost:8501 and select the sample you want to run from the s
 
 ![Object detection](../../docs/images/99_streamlit_examples.objectdetection.png)
 
+#### 11. Pose estimation
+
+![Pose estimation](../../docs/images/99_streamlit_examples.poseestimation.png)
+
 ## References
 
 - [🎈 Streamlit + LLM Examples App](https://github.com/streamlit/llm-examples)

apps/99_streamlit_examples/pages/11_Pose_estimation.py

@@ -0,0 +1,99 @@
+from io import BufferedReader, BytesIO
+
+import cv2
+import mediapipe as mp
+import numpy as np
+import streamlit as st
+from dotenv import load_dotenv
+
+load_dotenv()
+
+with st.sidebar:
+    # https://chuoling.github.io/mediapipe/solutions/pose.html#min_detection_confidence
+    min_detection_confidence = st.number_input(
+        label="Minimum detection confidence",
+        min_value=0.0,
+        max_value=1.0,
+        value=0.5,
+        key="min_detection_confidence",
+    )
+    # https://chuoling.github.io/mediapipe/solutions/pose.html#model_complexity
+    model_complexity = st.selectbox(
+        label="Model complexity",
+        options=[
+            0,
+            1,
+            2,
+        ],
+        index=1,
+        key="model_complexity",
+    )
+
+
+st.title("Pose estimation")
+
+st.info("Upload an image and AI will estimate the pose.")
+
+uploaded_file = st.file_uploader(
+    "Upload an image",
+    type=(
+        "jpg",
+        "jpeg",
+        "png",
+        "gif",
+        "bmp",
+        "tiff",
+    ),
+)
+if uploaded_file is not None:
+    button = st.button("Estimate pose")
+
+row1_left, row1_right = st.columns(2)
+with row1_left:
+    if uploaded_file:
+        st.image(
+            uploaded_file,
+            use_column_width=True,
+            caption="Input image",
+        )
+
+with row1_right:
+    if uploaded_file and button:
+        with st.spinner("Thinking..."):
+            mp_pose = mp.solutions.pose
+            mp_drawing = mp.solutions.drawing_utils
+
+            pose = mp_pose.Pose(
+                static_image_mode=True,
+                min_detection_confidence=min_detection_confidence,
+                model_complexity=model_complexity,
+            )
+
+            bytes_data = uploaded_file.getvalue()
+            cv2_img = cv2.imdecode(np.frombuffer(bytes_data, np.uint8), cv2.IMREAD_COLOR)
+            img = cv2.cvtColor(cv2_img, cv2.COLOR_BGR2RGB)
+            results = pose.process(img)
+            output_img = img.copy()
+            mp_drawing.draw_landmarks(
+                output_img,
+                results.pose_landmarks,
+                mp_pose.POSE_CONNECTIONS,
+                landmark_drawing_spec=mp_drawing.DrawingSpec(color=(255, 0, 0), thickness=2, circle_radius=10),
+                connection_drawing_spec=mp_drawing.DrawingSpec(color=(255, 0, 0), thickness=10),
+            )
+            ret, enco_img = cv2.imencode(".png", cv2.cvtColor(output_img, cv2.COLOR_BGR2RGB))
+            BytesIO_img = BytesIO(enco_img.tostring())
+            BufferedReader_img = BufferedReader(BytesIO_img)
+
+            # Output
+            st.image(
+                image=output_img,
+                use_column_width=True,
+                caption="Output image",
+            )
+            st.download_button(
+                label="Download",
+                data=BufferedReader_img,
+                file_name="output.png",
+                mime="image/png",
+            )