...

test.py

@@ -4,6 +4,7 @@ from diffsynth.pipelines.wan_video_new import WanVideoPipeline, ModelConfig
 from diffsynth.controlnets.processors import Annotator
 from modelscope import snapshot_download
 from tqdm import tqdm
+from PIL import Image
 
 
 # Load models
@@ -11,18 +12,29 @@ pipe = WanVideoPipeline.from_pretrained(
     torch_dtype=torch.bfloat16,
     device="cuda",
     model_configs=[
-        ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
+        ModelConfig(model_id="PAI/Wan2.1-Fun-V1.1-14B-Control", origin_file_pattern="diffusion_pytorch_model*.safetensors", offload_device="cpu"),
         ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="models_t5_umt5-xxl-enc-bf16.pth", offload_device="cpu"),
         ModelConfig(model_id="Wan-AI/Wan2.1-T2V-1.3B", origin_file_pattern="Wan2.1_VAE.pth", offload_device="cpu"),
-        # ModelConfig(model_id="Wan-AI/Wan2.1-I2V-14B-480P", origin_file_pattern="models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", offload_device="cpu"),
+        ModelConfig(model_id="Wan-AI/Wan2.1-I2V-14B-480P", origin_file_pattern="models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth", offload_device="cpu"),
     ],
 )
-pipe.enable_vram_management(num_persistent_param_in_dit=0)
+pipe.enable_vram_management(num_persistent_param_in_dit=10*10**9)
+
+
+video = VideoData(rf"D:\pr_projects\20250503_dance\data\双马尾竖屏暴击！你的微笑就是彩虹的微笑♥ - 1.双马尾竖屏暴击！你的微笑就是彩虹的微笑♥(Av114086629088385,P1).mp4", height=832, width=480)
+annotator = Annotator("openpose")
+video = [video[i] for i in tqdm(range(450, 450+1*17, 1))]
+save_video(video, "video_input.mp4", fps=60, quality=5)
+control_video = [annotator(f)  for f in tqdm(video)]
+save_video(control_video, "video_control.mp4", fps=60, quality=5)
+reference_image = Image.open(rf"D:\pr_projects\20250503_dance\data\marmot.png").resize((480, 832))
 
-# Text-to-video
 video = pipe(
-    prompt="纪实摄影风格画面，一只活泼的小狗在绿茵茵的草地上迅速奔跑。小狗毛色棕黄，两只耳朵立起，神情专注而欢快。阳光洒在它身上，使得毛发看上去格外柔软而闪亮。背景是一片开阔的草地，偶尔点缀着几朵野花，远处隐约可见蓝天和几片白云。透视感鲜明，捕捉小狗奔跑时的动感和四周草地的生机。中景侧面移动视角。",
+    prompt="微距摄影风格特写画面，一只憨态可掬的土拨鼠正用后腿站立在碎石堆上，它在挥舞着双臂。金棕色的绒毛在阳光下泛着丝绸般的光泽，腹部毛发呈现浅杏色渐变，每根毛尖都闪烁着细密的光晕。两只黑曜石般的眼睛透出机警而温顺的光芒，鼻梁两侧的白色触须微微颤动，捕捉着空气中的气息。背景是虚化的灰绿色渐变，几簇嫩绿苔藓从画面右下角探出头来，与前景散落的鹅卵石形成微妙的景深对比。土拨鼠圆润的身形在逆光中勾勒出柔和的轮廓，耳朵紧贴头部的姿态流露出戒备中的天真，整个画面洋溢着自然界生灵特有的灵动与纯真。",
     negative_prompt="色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走",
     seed=0, tiled=True,
+    height=832, width=480, num_frames=len(control_video),
+    control_video=control_video, reference_image=reference_image,
+    # num_inference_steps=30, cfg_scale=1,
 )
-save_video(video, "video1.mp4", fps=15, quality=5)
+save_video(video, "video1.mp4", fps=60, quality=5)