feature: added rendering

app.py

@@ -1,10 +1,19 @@
-import streamlit as st
-import streamlit.components.v1 as components
 import subprocess
 import os
 import glob
+import streamlit as st
+import streamlit.components.v1 as components
+import base64
+
+# Function to find the latest file with a given extension in a specified directory
+def find_latest_file(base_path, extension):
+    list_of_files = glob.glob(f'{base_path}/*.{extension}')
+    if not list_of_files:
+        return None
+    latest_file = max(list_of_files, key=os.path.getctime)
+    return latest_file
 
-# Function to run the model and generate HTML
+# Function to run the generate.py script and return paths of generated HTML and NPY files
 def generate_html(text_input, length):
     command = [
         "python", "generate.py",
@@ -15,18 +24,30 @@ def generate_html(text_input, length):
     ]
     try:
         result = subprocess.run(command, check=True, text=True, capture_output=True)
-        return find_latest_html_file('output')
+        html_file = find_latest_file('output', 'html')
+        npy_file = find_latest_file('output', 'npy')
+        return html_file, npy_file
     except subprocess.CalledProcessError as e:
-        print("Error:", e.stderr)  
-        return None
+        st.error(f"Error: {e.stderr}")  
+        return None, None
 
-# Function to find the latest HTML file in the specified directory
-def find_latest_html_file(base_path):
-    list_of_files = glob.glob(f'{base_path}/*.html')
-    if not list_of_files:
+# Function to run render_final.py script with the generated NPY file
+def run_render_final(npy_file_path):
+    command = ["python", "render_final.py", npy_file_path]
+    try:
+        gif_res = subprocess.run(command, check=True, text=True, capture_output=True)
+        gif_file_path = find_latest_file('output', 'gif')
+        return gif_file_path
+    except subprocess.CalledProcessError as e:
+        st.error(f"Error: {e.stderr}")
         return None
-    latest_file = max(list_of_files, key=os.path.getctime)
-    return latest_file
+
+# Function to convert GIF to base64
+def gif_to_base64(gif_file_path):
+    with open(gif_file_path, "rb") as gif_file:
+        gif_bytes = gif_file.read()
+    base64_gif = base64.b64encode(gif_bytes).decode("utf-8")
+    return base64_gif
 
 # Initialize session state 
 if 'text_input' not in st.session_state:
@@ -34,16 +55,14 @@ if 'text_input' not in st.session_state:
 if 'length' not in st.session_state:
     st.session_state.length = 156
 
-# handler to update session state and rerun the app
+# Handler to update session state and rerun the app
 def select_prompt(prompt, prompt_length):
     st.session_state.text_input = prompt
     st.session_state.length = prompt_length
 
-# app layout
-components.html("""
-    <h1 style='text-align: center; color: white;'>MMM Model Demo</h1>
-    """, height=100)
 
+# App layout
+components.html("<h1 style='text-align: center; color: white;'>MMM Model Demo</h1>", height=100)
 
 prompts = [
     ("A person walks forward then turns completely around and does a cartwheel", 196),
@@ -73,13 +92,46 @@ with input_placeholder.container():
     text_input = st.text_area("Enter text here:", value=st.session_state.text_input, key="text_input", height=300)
     length = st.number_input("Length of the generated text:", value=st.session_state.length, key="length")
 
-# Button trigger to generate HTML
-if st.button("Generate HTML"):
-    if st.session_state.text_input and st.session_state.length:
-        html_file_path = generate_html(st.session_state.text_input, st.session_state.length)
-        if html_file_path and os.path.exists(html_file_path):
-            with open(html_file_path, 'r') as file:
-                html_content = file.read()
-            components.html(html_content, height=800, width=800)
+# Place the buttons side by side
+button_col1, button_col2 = st.columns(2)
+
+with button_col1:
+    if st.button("Generate HTML"):
+        if st.session_state.text_input and st.session_state.length:
+            html_file_path, npy_file_path = generate_html(st.session_state.text_input, st.session_state.length)
+            if html_file_path and npy_file_path:
+                st.session_state.html_file_path = html_file_path
+                st.session_state.npy_file_path = npy_file_path
+                
+                # Display the HTML file content
+                with open(html_file_path, 'r') as file:
+                    html_content = file.read()
+                st.session_state.html_content = html_content
+            else:
+                st.error("Error generating files. Please try again.")
+
+with button_col2:
+    if st.button("Render Skeleton"):
+        if 'npy_file_path' in st.session_state and st.session_state.npy_file_path:
+            gif_file_path = run_render_final(st.session_state.npy_file_path)
+            if gif_file_path:
+                st.session_state.gif_file_path = gif_file_path
+                st.session_state.gif_base64 = gif_to_base64(gif_file_path)
         else:
-            st.error("Error generating HTML file. Please try again.")
+            st.error("No npy file found. Please generate HTML first.")
+
+
+# Display the results side by side using HTML components
+if 'html_content' in st.session_state or 'gif_base64' in st.session_state:
+    html_content = st.session_state.html_content if 'html_content' in st.session_state else ""
+    gif_base64 = st.session_state.gif_base64 if 'gif_base64' in st.session_state else ""
+    
+    disp_col1, disp_col2 = st.columns([1, 1])
+    
+    with disp_col1:
+        components.html(html_content, height=800, scrolling=True)
+    
+    with disp_col2:
+        if gif_base64:
+            gif_html = f'<img src="data:image/gif;base64,{gif_base64}" style="width:100%;">'
+            components.html(gif_html, height=800, scrolling=True)

generate.py

@@ -291,15 +291,22 @@ if __name__ == '__main__':
     pred_pose = mmm([args.text], torch.tensor([args.length]), rand_pos=False)
     num_joints = 22
     
-    pred_pose = pred_pose[:args.length, :].detach().cpu()
+    std = np.load('./exit/t2m-std.npy')
+    mean = np.load('./exit/t2m-mean.npy')
+    
+    norm_pose = pred_pose[0].detach().cpu().numpy() * std + mean
+    norm_pose = torch.tensor(norm_pose)
     
-    converted_pose = recover_from_ric(pred_pose[0].detach().cpu(), num_joints).unsqueeze(0).numpy()
+    trimmed_pose = norm_pose[:args.length, :].unsqueeze(0).float()
+    print(trimmed_pose.shape)
     
-    np.save('./output/mmm-pred.npy', converted_pose)
+    converted_pose = recover_from_ric(trimmed_pose[0].detach().cpu(), num_joints).unsqueeze(0).numpy()
+    print(converted_pose.shape)
+
+    filename = '_'.join(args.text.split(' '))+'_'+str(args.length)
+    np.save('./output/'+filename+'.npy', converted_pose)
     print('File saved successfully')
 
-    std = np.load('./exit/t2m-std.npy')
-    mean = np.load('./exit/t2m-mean.npy')
     file_name = '_'.join(args.text.split(' '))+'_'+str(args.length)
     visualize_2motions(pred_pose[0].detach().cpu().numpy(), std, mean, 't2m', args.length, save_path='./output/'+file_name+'.html')
 

render_final.py

@@ -164,8 +164,7 @@ def render(motions, outdir='test_vis', device_id=0, name=None, pred=True):
     gif_path = os.path.join(outdir, f'{name}.gif')
     imageio.mimsave(gif_path, out, fps=20)
     
-
-
+    
 
 
 if __name__ == "__main__":

requirements.txt

@@ -1,37 +1,101 @@
-git+https://github.com/openai/CLIP.git
-git+https://github.com/mmatl/pyopengl.git
+altair==5.3.0
+attrs==23.2.0
 beautifulsoup4==4.12.3
+blinker==1.8.2
+cachetools==5.3.3
+certifi==2024.6.2
+charset-normalizer==3.3.2
+chumpy==0.70
+click==8.1.7
+contourpy==1.1.1
+cycler==0.12.1
 einops==0.8.0
 fastjsonschema==2.20.0
+filelock==3.15.4
+fonttools==4.53.0
 freetype-py==2.4.0
 fsspec==2024.6.1
 ftfy==6.2.0
 gdown==5.2.0
+gitdb==4.0.11
+GitPython==3.1.43
 h5py==3.11.0
+idna==3.7
 imageio==2.34.2
+importlib_resources==6.4.0
+Jinja2==3.1.4
 jsonschema==4.22.0
 jsonschema-specifications==2023.12.1
 jupyter_core==5.7.2
+kiwisolver==1.4.5
 mapbox-earcut==1.0.1
+markdown-it-py==3.0.0
+MarkupSafe==2.1.5
+matplotlib==3.7.5
+mdurl==0.1.2
+mpmath==1.3.0
 nbformat==5.10.4
+networkx==3.1
 numpy==1.23.3
+nvidia-cublas-cu12==12.1.3.1
+nvidia-cuda-cupti-cu12==12.1.105
+nvidia-cuda-nvrtc-cu12==12.1.105
+nvidia-cuda-runtime-cu12==12.1.105
+nvidia-cudnn-cu12==8.9.2.26
+nvidia-cufft-cu12==11.0.2.54
+nvidia-curand-cu12==10.3.2.106
+nvidia-cusolver-cu12==11.4.5.107
+nvidia-cusparse-cu12==12.1.0.106
+nvidia-nccl-cu12==2.20.5
+nvidia-nvjitlink-cu12==12.5.40
+nvidia-nvtx-cu12==12.1.105
+oauthlib==3.2.2
+packaging==24.1
+pandas==2.0.3
+pillow==10.3.0
 pkgutil_resolve_name==1.3.10
-plotly
+platformdirs==4.2.2
+plotly==5.22.0
 protobuf==3.20.3
+pyarrow==16.1.0
+pyasn1==0.4.8
 pyasn1-modules==0.2.8
+pydeck==0.9.1
 pyglet==2.0.15
+Pygments==2.18.0
+pyparsing==3.1.2
 pyrender==0.1.45
+PySocks==1.7.1
+python-dateutil==2.9.0.post0
+pytz==2024.1
 referencing==0.35.1
 regex==2024.5.15
+requests==2.32.3
 requests-oauthlib==1.3.0
+rich==13.7.1
 rpds-py==0.18.1
 scipy==1.10.1
 shapely==2.0.4
+six==1.16.0
+smmap==5.0.1
+smplx==0.1.28
 soupsieve==2.5
+streamlit==1.36.0
+sympy==1.12.1
+tenacity==8.4.2
+toml==0.10.2
+toolz==0.12.1
 torch==2.3.1
 torchaudio==2.3.1
 torchvision==0.18.1
+tornado==6.4.1
 tqdm==4.66.4
 traitlets==5.14.3
 trimesh==4.4.1
+triton==2.3.1
+typing_extensions==4.12.2
+tzdata==2024.1
+urllib3==2.2.2
+watchdog==4.0.1
 wcwidth==0.2.13
+zipp==3.19.2