Finish merge

app_v2v.py

@@ -50,8 +50,6 @@ if torch.cuda.device_count() > 0:
     print(f'Free VRAM {free_mem_gb} GB')
     print(f'High-VRAM Mode: {high_vram}')
 
-
-
     text_encoder = LlamaModel.from_pretrained("hunyuanvideo-community/HunyuanVideo", subfolder='text_encoder', torch_dtype=torch.float16).cpu()
     text_encoder_2 = CLIPTextModel.from_pretrained("hunyuanvideo-community/HunyuanVideo", subfolder='text_encoder_2', torch_dtype=torch.float16).cpu()
     tokenizer = LlamaTokenizerFast.from_pretrained("hunyuanvideo-community/HunyuanVideo", subfolder='tokenizer')
@@ -926,7 +924,8 @@ adapted from the official code repo [FramePack](https://github.com/lllyasviel/Fr
             t2v = gr.Checkbox(label="Do text-to-video (ignored for video extension)", value=False)
 
             with gr.Row():
-                start_button = gr.Button(value="Start Generation", variant="primary")
+                start_button = gr.Button(value="Generate from image", variant="primary")
+                start_button_video = gr.Button(value="Generate from video", variant="primary")
                 end_button = gr.Button(value="End Generation", variant="stop", interactive=False)
 
             total_second_length = gr.Slider(label="Video Length to Generate (seconds)", minimum=1, maximum=120, value=2, step=0.1)
@@ -983,7 +982,8 @@ adapted from the official code repo [FramePack](https://github.com/lllyasviel/Fr
     # 20250506 pftq: Updated inputs to include num_clean_frames
     ips = [input_image, prompt, t2v, n_prompt, randomize_seed, seed, total_second_length, latent_window_size, steps, cfg, gs, rs, gpu_memory_preservation, use_teacache, mp4_crf]
     ips_video = [input_video, prompt, n_prompt, randomize_seed, seed, batch, resolution, total_second_length, latent_window_size, steps, cfg, gs, rs, gpu_memory_preservation, use_teacache, no_resize, mp4_crf, num_clean_frames, vae_batch]
-    start_button.click(fn=process_video, inputs=ips_video, outputs=[result_video, preview_image, progress_desc, progress_bar, start_button, end_button])
+    start_button.click(fn=process, inputs=ips, outputs=[result_video, preview_image, progress_desc, progress_bar, start_button, end_button])
+    start_button_video.click(fn=process_video, inputs=ips_video, outputs=[result_video, preview_image, progress_desc, progress_bar, start_button_video, end_button])
     end_button.click(fn=end_process)
 
     with gr.Row(elem_id="image_examples", visible=False):
@@ -1093,7 +1093,7 @@ adapted from the official code repo [FramePack](https://github.com/lllyasviel/Fr
         run_on_click = True,
         fn = process_video,
 	    inputs = ips_video,
-	    outputs = [result_video, preview_image, progress_desc, progress_bar, start_button, end_button],
+	    outputs = [result_video, preview_image, progress_desc, progress_bar, start_button_video, end_button],
         cache_examples = True,
     )
 
@@ -1114,7 +1114,7 @@ adapted from the official code repo [FramePack](https://github.com/lllyasviel/Fr
         inputs=[input_image_debug, input_video_debug, prompt_debug, total_second_length_debug],
         outputs=[]
     )
-    
+
     input_video_debug.upload(
         fn=handle_field_debug_change,
         inputs=[input_image_debug, input_video_debug, prompt_debug, total_second_length_debug],