Update app.py

app.py

@@ -84,7 +84,7 @@ def image_to_3d(
     slat_guidance_strength: float,
     slat_sampling_steps: int,
     req: gr.Request,
-) -> Tuple[dict, str, str]:
+) -> Tuple[dict, str, str, str]:
     """
     Convert an image to a 3D model.
     """
@@ -103,20 +103,19 @@ def image_to_3d(
             "cfg_strength": slat_guidance_strength,
         },
     )
-
-    # Generate video preview
+    
     video = render_utils.render_video(outputs['gaussian'][0], num_frames=120)['color']
     video_geo = render_utils.render_video(outputs['mesh'][0], num_frames=120)['normal']
     video = [np.concatenate([video[i], video_geo[i]], axis=1) for i in range(len(video))]
     trial_id = str(uuid.uuid4())
     video_path = os.path.join(user_dir, f"{trial_id}.mp4")
     imageio.mimsave(video_path, video, fps=15)
-
-    # Save full quality GLB
+    
+    # Generate full quality GLB
     glb = postprocessing_utils.to_glb(
         outputs['gaussian'][0], 
         outputs['mesh'][0],
-        simplify=0.0,  # No simplification for full quality
+        simplify=0.0,  # No simplification
         fill_holes=True,
         fill_holes_max_size=0.04,
         texture_size=2048,  # Maximum texture size
@@ -125,45 +124,39 @@ def image_to_3d(
     full_glb_path = os.path.join(user_dir, f"{trial_id}_full.glb")
     glb.export(full_glb_path)
     
-    # Pack state for potential reduced version
     state = pack_state(outputs['gaussian'][0], outputs['mesh'][0], trial_id)
-    
-    return state, video_path, full_glb_path
+    return state, video_path, model_output, full_glb_path
 
-def extract_reduced_glb(
+def extract_glb(
     state: dict,
     mesh_simplify: float,
     texture_size: int,
     req: gr.Request,
 ) -> Tuple[str, str]:
     """
-    Extract a reduced quality GLB file.
+    Extract a reduced GLB file from the 3D model.
     """
     user_dir = os.path.join(TMP_DIR, str(req.session_hash))
     gs, mesh, trial_id = unpack_state(state)
-    
-    # Create reduced quality GLB with user settings
     glb = postprocessing_utils.to_glb(
-        gs, 
-        mesh,
+        gs, mesh,
         simplify=mesh_simplify,
         fill_holes=True,
         fill_holes_max_size=0.04,
         texture_size=texture_size,
         verbose=False
     )
-    reduced_glb_path = os.path.join(user_dir, f"{trial_id}_reduced.glb")
-    glb.export(reduced_glb_path)
-    
-    return reduced_glb_path, reduced_glb_path
+    glb_path = os.path.join(user_dir, f"{trial_id}_reduced.glb")
+    glb.export(glb_path)
+    return glb_path, glb_path
 
 with gr.Blocks(delete_cache=(600, 600)) as demo:
     gr.Markdown("""
     ## Image to 3D Asset with [TRELLIS](https://trellis3d.github.io/)
     * Upload an image and click "Generate" to create a 3D asset
-    * After generation:
-        * Download the full quality GLB (no mesh simplification, maximum texture resolution)
-        * Or create a reduced size version with customizable settings
+    * After generation, you can:
+        * Download the full quality GLB immediately
+        * Create a reduced size version with the extraction settings below
     """)
     
     with gr.Row():
@@ -184,17 +177,15 @@ with gr.Blocks(delete_cache=(600, 600)) as demo:
 
             generate_btn = gr.Button("Generate")
             
-            with gr.Accordion(label="Reduced GLB Settings", open=False):
-                mesh_simplify = gr.Slider(0.0, 0.98, label="Mesh Simplification", value=0.95, step=0.01, 
-                                        info="Higher values = more reduction")
+            with gr.Accordion(label="GLB Extraction Settings", open=False):
+                mesh_simplify = gr.Slider(0.0, 0.98, label="Simplify", value=0.95, step=0.01)
                 texture_size = gr.Slider(512, 2048, label="Texture Size", value=1024, step=512)
             
-            extract_reduced_btn = gr.Button("Extract Reduced GLB", interactive=False)
+            extract_glb_btn = gr.Button("Extract Reduced GLB", interactive=False)
 
         with gr.Column():
             video_output = gr.Video(label="Generated 3D Asset", autoplay=True, loop=True, height=300)
             model_output = LitModel3D(label="3D Model Preview", exposure=20.0, height=300)
-            gr.Markdown("### Download Options")
             with gr.Row():
                 download_full = gr.DownloadButton(label="Download Full-Quality GLB", interactive=False)
                 download_reduced = gr.DownloadButton(label="Download Reduced GLB", interactive=False)
@@ -230,4 +221,28 @@ with gr.Blocks(delete_cache=(600, 600)) as demo:
         inputs=[randomize_seed, seed],
         outputs=[seed],
     ).then(
-        image_to_3d,
+        image_to_3d,
+        inputs=[image_prompt, seed, ss_guidance_strength, ss_sampling_steps, slat_guidance_strength, slat_sampling_steps],
+        outputs=[output_buf, video_output, model_output, download_full],
+    ).then(
+        lambda: [gr.Button(interactive=True), gr.Button(interactive=True), gr.Button(interactive=False)],
+        outputs=[download_full, extract_glb_btn, download_reduced],
+    )
+
+    extract_glb_btn.click(
+        extract_glb,
+        inputs=[output_buf, mesh_simplify, texture_size],
+        outputs=[model_output, download_reduced],
+    ).then(
+        lambda: gr.Button(interactive=True),
+        outputs=[download_reduced],
+    )
+
+if __name__ == "__main__":
+    pipeline = TrellisImageTo3DPipeline.from_pretrained("JeffreyXiang/TRELLIS-image-large")
+    pipeline.cuda()
+    try:
+        pipeline.preprocess_image(Image.fromarray(np.zeros((512, 512, 3), dtype=np.uint8)))
+    except:
+        pass
+    demo.launch()