update

app.py

@@ -1,7 +1,7 @@
 import os
 import torch
 import sys
-import spaces
+import spaces #fixme
 
 
 import gradio as gr
@@ -9,12 +9,20 @@ import random
 from configs.infer_config import get_parser
 from huggingface_hub import hf_hub_download
 
-i2v_examples = [
-    ['test/images/boy.png', 0, 1.0, '0 40', '0 0', '0 0',  50, 123],
-    ['test/images/car.jpeg', 0, 1.0, '0 -35', '0 0', '0 -0.1',  50, 123],
-    ['test/images/fruit.jpg', 0, 1.0, '0 -3 -15 -20 -17 -5 0', '0 -2 -5 -10 -8 -5 0 2 5 3 0', '0 0',  50, 123],
-    ['test/images/room.png', 5, 1.0, '0 3 10 20 17 10 0', '0 -2 -8 -6 0 2 5 3 0', '0 -0.02 -0.09 -0.16 -0.09 0',  50, 123],
-    ['test/images/castle.png', 0, 1.0, '0 30', '0 -1 -5 -4 0 1 5 4 0', '0 -0.2',  50, 123],
+traj_examples = [
+    ['0 40', '0 0', '0 0'],
+    ['0 -35', '0 0', '0 -0.1'],
+    ['0 -3 -15 -20 -17 -5 0', '0 -2 -5 -10 -8 -5 0 2 5 3 0', '0 0'],
+    ['0 3 10 20 17 10 0', '0 -2 -8 -6 0 2 5 3 0', '0 -0.02 -0.09 -0.16 -0.09 0'],
+    ['0 30', '0 -1 -5 -4 0 1 5 4 0', '0 -0.2'],
+]
+
+img_examples = [
+    ['test/images/boy.png',0,1],
+    ['test/images/car.jpeg',5,1],
+    ['test/images/fruit.jpg',5,1],
+    ['test/images/room.png',10,1],
+    ['test/images/castle.png',-4,1],
 ]
 
 max_seed = 2 ** 31
@@ -27,7 +35,7 @@ def download_model():
         if not os.path.exists(local_file):
             hf_hub_download(repo_id=REPO_ID, filename=filename, local_dir='./checkpoints/', force_download=True)
     
-download_model()
+download_model() #fixme
 parser = get_parser() # infer_config.py
 opts = parser.parse_args() # default device: 'cuda:0'
 opts.save_dir = './'
@@ -35,7 +43,7 @@ os.makedirs(opts.save_dir,exist_ok=True)
 test_tensor = torch.Tensor([0]).cuda()
 opts.device = str(test_tensor.device)
 
-# install pytorch3d
+# install pytorch3d # fixme
 pyt_version_str=torch.__version__.split("+")[0].replace(".", "")
 version_str="".join([
     f"py3{sys.version_info.minor}_cu",
@@ -50,10 +58,12 @@ print(f'>>> System info: {version_str}')
 
 from viewcrafter import ViewCrafter
 
+
 def viewcrafter_demo(opts):
     css = """#input_img {max-width: 1024px !important} #output_vid {max-width: 1024px; max-height:576px} #random_button {max-width: 100px !important}"""
     image2video = ViewCrafter(opts, gradio = True)
-    image2video.run_gradio = spaces.GPU(image2video.run_gradio, duration=300)
+    image2video.run_traj = spaces.GPU(image2video.run_traj, duration=50) # fixme
+    image2video.run_gen = spaces.GPU(image2video.run_gen, duration=250) # fixme
     with gr.Blocks(analytics_enabled=False, css=css) as viewcrafter_iface:
         gr.Markdown("<div align='center'> <h1> ViewCrafter: Taming Video Diffusion Models for High-fidelity Novel View Synthesis </span> </h1> \
                       <h2 style='font-weight: 450; font-size: 1rem; margin: 0rem'>\
@@ -65,58 +75,78 @@ def viewcrafter_demo(opts):
                         <a href='https://scholar.google.com/citations?hl=en&user=4oXBp9UAAAAJ&view_op=list_works&sortby=pubdate/'>Ying Shan</a>\
                         <a href=''>Yonghong Tian</a>\
                     </h2> \
-                     <a style='font-size:18px;color: #FF5DB0' href='https://github.com/Drexubery/ViewCrafter/blob/main/docs/render_help.md'> [Guideline] </a>\
                      <a style='font-size:18px;color: #000000' href=''> [ArXiv] </a>\
                      <a style='font-size:18px;color: #000000' href='https://drexubery.github.io/ViewCrafter/'> [Project Page] </a>\
-                     <a style='font-size:18px;color: #000000' href='https://github.com/Drexubery/ViewCrafter'> [Github] </a> </div>") 
-                
-        #######image2video######
-        with gr.Tab(label="ViewCrafter_25, 'single_view_txt' mode"):
-            with gr.Column():
-                with gr.Row():
-                    with gr.Column():
-                        with gr.Row():
-                            i2v_input_image = gr.Image(label="Input Image",elem_id="input_img")
-                        with gr.Row():
-                            i2v_elevation = gr.Slider(minimum=-45, maximum=45, step=1, elem_id="elevation", label="elevation", value=5)
-                        with gr.Row():
-                            i2v_center_scale = gr.Slider(minimum=0.1, maximum=2, step=0.1, elem_id="i2v_center_scale", label="center_scale", value=1)
-                        with gr.Row():
-                            i2v_d_phi = gr.Text(label='d_phi sequence, should start with 0')
-                        with gr.Row():
-                            i2v_d_theta = gr.Text(label='d_theta sequence, should start with 0')
-                        with gr.Row():
-                            i2v_d_r = gr.Text(label='d_r sequence, should start with 0')
-                        with gr.Row():
-                            i2v_steps = gr.Slider(minimum=1, maximum=50, step=1, elem_id="i2v_steps", label="Sampling steps", value=50)
-                        with gr.Row():
-                            i2v_seed = gr.Slider(label='Random Seed', minimum=0, maximum=max_seed, step=1, value=123)
-                        i2v_end_btn = gr.Button("Generate")
-                    # with gr.Tab(label='Result'):
-                    with gr.Column():
-                        with gr.Row():
-                            i2v_traj_video = gr.Video(label="Camera Trajectory",elem_id="traj_vid",autoplay=True,show_share_button=True)
-                        with gr.Row():
-                            i2v_output_video = gr.Video(label="Generated Video",elem_id="output_vid",autoplay=True,show_share_button=True)
-
-                gr.Examples(examples=i2v_examples,
-                            inputs=[i2v_input_image, i2v_elevation, i2v_center_scale, i2v_d_phi, i2v_d_theta, i2v_d_r, i2v_steps, i2v_seed],
-                            outputs=[i2v_traj_video,i2v_output_video],
-                            fn = image2video.run_gradio,
-                            cache_examples='lazy',
-                )
-
-            # image2video.run_gradio(i2v_input_image='test/images/boy.png', i2v_elevation='10', i2v_d_phi='0 40', i2v_d_theta='0 0', i2v_d_r='0 0', i2v_center_scale=1, i2v_steps=50, i2v_seed=123)
-            i2v_end_btn.click(inputs=[i2v_input_image, i2v_elevation, i2v_center_scale, i2v_d_phi, i2v_d_theta, i2v_d_r, i2v_steps, i2v_seed],
-                            outputs=[i2v_traj_video,i2v_output_video],
-                            fn = image2video.run_gradio
+                     <a style='font-size:18px;color: #FF5DB0' href='https://github.com/Drexubery/ViewCrafter'> [Github] </a>\
+                     <a style='font-size:18px;color: #000000' href='https://www.youtube.com/watch?v=WGIEmu9eXmU'> [Video] </a> </div>")
+
+
+        with gr.Column():
+            # step 0: tutorial
+            gr.Markdown("## Step 0: Read tutorial", show_label=False)
+            gr.Markdown("<div align='left' style='font-size:18px;color: #000000'>Please refer to the tutorial <a href='https://github.com/TencentARC/MotionCtrl/blob/svd/doc/tutorial.md' target='_blank'>here</a> for best practice, which includes the cameara system defination and the renderer parameters.</div>")
+
+            # step 2: input an image
+            gr.Markdown("---\n## Step 1: Input an Image, selet an elevation angle and a center_scale factor", show_label=False, visible=True)
+            gr.Markdown("<div align='left' style='font-size:18px;color: #000000'>1. Estimate an elevation angle  that represents the angle at which the image was taken; a value bigger than 0 indicates a top-down view, and it doesn't need to be precise. <br>2. The origin of the world coordinate system is by default defined at the point cloud corresponding to the center pixel of the input image. You can adjust the position of the origin by modifying center_scale; a value smaller than 1 brings the origin closer to you.</div>")
+            with gr.Row(equal_height=True):
+                with gr.Column(scale=2):
+                    with gr.Row():
+                        i2v_input_image = gr.Image(label="Input Image",elem_id="input_img")
+                    with gr.Row():
+                        i2v_elevation = gr.Slider(minimum=-45, maximum=45, step=1, elem_id="elevation", label="elevation", value=5)
+                        i2v_center_scale = gr.Slider(minimum=0.1, maximum=2, step=0.1, elem_id="i2v_center_scale", label="center_scale", value=1)
+            gr.Examples(examples=img_examples,
+                        inputs=[i2v_input_image,i2v_elevation,i2v_center_scale],
+                        examples_per_page=6
             )
+            # step 2 - camera trajectory generation
+            gr.Markdown("---\n## Step 2: Input camera trajectory", show_label=False, visible=True)
+            gr.Markdown("<div align='left' style='font-size:18px;color: #000000'> Input a d_phi sequence, a d_theta sequence, and a d_r sequence to generate a camera trajectory. In the sequences, a positive d_phi moves the camera to the right, a negative d_theta moves the camera up, and a negative d_r moves the camera forward. Ensure that each sequence starts with 0 and contains at least two elements (a start and an end). </div>")
+            with gr.Row():
+                with gr.Column():
+                    # camera_mode = gr.Radio(choices=CAMERA_MOTION_MODE, value=CAMERA_MOTION_MODE[0], label="Camera Motion Control Mode", interactive=True, visible=False)
+                    i2v_d_phi = gr.Text(label='d_phi sequence')
+                    i2v_d_theta = gr.Text(label='d_theta sequence')
+                    i2v_d_r = gr.Text(label='d_r sequence')
+                    i2v_start_btn = gr.Button("Generate trajectory")
+                    # camera_info = gr.Button(value="Proceed", visible=False)      
+                with gr.Column():
+                    i2v_traj_video = gr.Video(label="Camera Trajectory",elem_id="traj_vid",autoplay=True,show_share_button=True)          
+            
+            gr.Examples(examples=traj_examples,
+                        inputs=[i2v_d_phi, i2v_d_theta, i2v_d_r],
+            )
+
+            # step 3 - Generate video
+            gr.Markdown("---\n## Step 3: Generate video", show_label=False, visible=True)
+            gr.Markdown("<div align='left' style='font-size:18px;color: #000000'> You can reduce the sampling steps for faster inference; try different random seed if the result is not satisfying. </div>")
+            with gr.Row():
+                with gr.Column():
+                    i2v_steps = gr.Slider(minimum=1, maximum=50, step=1, elem_id="i2v_steps", label="Sampling steps", value=50)
+                    i2v_seed = gr.Slider(label='Random seed', minimum=0, maximum=max_seed, step=1, value=123)
+                    i2v_end_btn = gr.Button("Generate video")
+                    # with gr.Tab(label='Result'):
+                with gr.Column():
+                    i2v_output_video = gr.Video(label="Generated Video",elem_id="output_vid",autoplay=True,show_share_button=True)
+
+
+
+        i2v_start_btn.click(inputs=[i2v_input_image, i2v_elevation, i2v_center_scale, i2v_d_phi, i2v_d_theta, i2v_d_r],
+                        outputs=[i2v_traj_video],
+                        fn = image2video.run_traj
+        )        
+        
+        i2v_end_btn.click(inputs=[i2v_steps, i2v_seed],
+                        outputs=[i2v_output_video],
+                        fn = image2video.run_gen
+        )
 
     return viewcrafter_iface
 
 
 viewcrafter_iface = viewcrafter_demo(opts)
 viewcrafter_iface.queue(max_size=10)
-viewcrafter_iface.launch()
-# viewcrafter_iface.launch(server_name='127.0.0.1', server_port=80, max_threads=1,debug=False)
+viewcrafter_iface.launch() #fixme
+# viewcrafter_iface.launch(server_name='11.220.92.96', server_port=80, max_threads=1,debug=False)
 

app_old.py

@@ -0,0 +1,122 @@
+import os
+import torch
+import sys
+import spaces
+
+
+import gradio as gr
+import random
+from configs.infer_config import get_parser
+from huggingface_hub import hf_hub_download
+
+i2v_examples = [
+    ['test/images/boy.png', 0, 1.0, '0 40', '0 0', '0 0',  50, 123],
+    ['test/images/car.jpeg', 0, 1.0, '0 -35', '0 0', '0 -0.1',  50, 123],
+    ['test/images/fruit.jpg', 0, 1.0, '0 -3 -15 -20 -17 -5 0', '0 -2 -5 -10 -8 -5 0 2 5 3 0', '0 0',  50, 123],
+    ['test/images/room.png', 5, 1.0, '0 3 10 20 17 10 0', '0 -2 -8 -6 0 2 5 3 0', '0 -0.02 -0.09 -0.16 -0.09 0',  50, 123],
+    ['test/images/castle.png', 0, 1.0, '0 30', '0 -1 -5 -4 0 1 5 4 0', '0 -0.2',  50, 123],
+]
+
+max_seed = 2 ** 31
+
+def download_model():
+    REPO_ID = 'Drexubery/ViewCrafter_25'
+    filename_list = ['model.ckpt']
+    for filename in filename_list:
+        local_file = os.path.join('./checkpoints/', filename)
+        if not os.path.exists(local_file):
+            hf_hub_download(repo_id=REPO_ID, filename=filename, local_dir='./checkpoints/', force_download=True)
+    
+download_model()
+parser = get_parser() # infer_config.py
+opts = parser.parse_args() # default device: 'cuda:0'
+opts.save_dir = './'
+os.makedirs(opts.save_dir,exist_ok=True)
+test_tensor = torch.Tensor([0]).cuda()
+opts.device = str(test_tensor.device)
+
+# install pytorch3d
+pyt_version_str=torch.__version__.split("+")[0].replace(".", "")
+version_str="".join([
+    f"py3{sys.version_info.minor}_cu",
+    torch.version.cuda.replace(".",""),
+    f"_pyt{pyt_version_str}"
+])
+print(version_str)
+os.system(f"{sys.executable} -m pip install --no-index --no-cache-dir pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html")
+os.system("mkdir -p checkpoints/ && wget https://download.europe.naverlabs.com/ComputerVision/DUSt3R/DUSt3R_ViTLarge_BaseDecoder_512_dpt.pth -P checkpoints/")
+print(f'>>> System info: {version_str}')
+
+
+from viewcrafter import ViewCrafter
+
+def viewcrafter_demo(opts):
+    css = """#input_img {max-width: 1024px !important} #output_vid {max-width: 1024px; max-height:576px} #random_button {max-width: 100px !important}"""
+    image2video = ViewCrafter(opts, gradio = True)
+    image2video.run_gradio = spaces.GPU(image2video.run_gradio, duration=300)
+    with gr.Blocks(analytics_enabled=False, css=css) as viewcrafter_iface:
+        gr.Markdown("<div align='center'> <h1> ViewCrafter: Taming Video Diffusion Models for High-fidelity Novel View Synthesis </span> </h1> \
+                      <h2 style='font-weight: 450; font-size: 1rem; margin: 0rem'>\
+                        <a href='https://scholar.google.com/citations?user=UOE8-qsAAAAJ&hl=zh-CN'>Wangbo Yu</a>, \
+                        <a href='https://doubiiu.github.io/'>Jinbo Xing</a>, <a href=''>Li Yuan</a>, \
+                        <a href='https://wbhu.github.io/'>Wenbo Hu</a>, <a href='https://xiaoyu258.github.io/'>Xiaoyu Li</a>,\
+                        <a href=''>Zhipeng Huang</a>, <a href='https://scholar.google.com/citations?user=qgdesEcAAAAJ&hl=en/'>Xiangjun Gao</a>,\
+                        <a href='https://www.cse.cuhk.edu.hk/~ttwong/myself.html/'>Tien-Tsin Wong</a>,\
+                        <a href='https://scholar.google.com/citations?hl=en&user=4oXBp9UAAAAJ&view_op=list_works&sortby=pubdate/'>Ying Shan</a>\
+                        <a href=''>Yonghong Tian</a>\
+                    </h2> \
+                     <a style='font-size:18px;color: #FF5DB0' href='https://github.com/Drexubery/ViewCrafter/blob/main/docs/render_help.md'> [Guideline] </a>\
+                     <a style='font-size:18px;color: #000000' href=''> [ArXiv] </a>\
+                     <a style='font-size:18px;color: #000000' href='https://drexubery.github.io/ViewCrafter/'> [Project Page] </a>\
+                     <a style='font-size:18px;color: #000000' href='https://github.com/Drexubery/ViewCrafter'> [Github] </a> </div>") 
+                
+        #######image2video######
+        with gr.Tab(label="ViewCrafter_25, 'single_view_txt' mode"):
+            with gr.Column():
+                with gr.Row():
+                    with gr.Column():
+                        with gr.Row():
+                            i2v_input_image = gr.Image(label="Input Image",elem_id="input_img")
+                        with gr.Row():
+                            i2v_elevation = gr.Slider(minimum=-45, maximum=45, step=1, elem_id="elevation", label="elevation", value=5)
+                        with gr.Row():
+                            i2v_center_scale = gr.Slider(minimum=0.1, maximum=2, step=0.1, elem_id="i2v_center_scale", label="center_scale", value=1)
+                        with gr.Row():
+                            i2v_d_phi = gr.Text(label='d_phi sequence, should start with 0')
+                        with gr.Row():
+                            i2v_d_theta = gr.Text(label='d_theta sequence, should start with 0')
+                        with gr.Row():
+                            i2v_d_r = gr.Text(label='d_r sequence, should start with 0')
+                        with gr.Row():
+                            i2v_steps = gr.Slider(minimum=1, maximum=50, step=1, elem_id="i2v_steps", label="Sampling steps", value=50)
+                        with gr.Row():
+                            i2v_seed = gr.Slider(label='Random Seed', minimum=0, maximum=max_seed, step=1, value=123)
+                        i2v_end_btn = gr.Button("Generate")
+                    # with gr.Tab(label='Result'):
+                    with gr.Column():
+                        with gr.Row():
+                            i2v_traj_video = gr.Video(label="Camera Trajectory",elem_id="traj_vid",autoplay=True,show_share_button=True)
+                        with gr.Row():
+                            i2v_output_video = gr.Video(label="Generated Video",elem_id="output_vid",autoplay=True,show_share_button=True)
+
+                gr.Examples(examples=i2v_examples,
+                            inputs=[i2v_input_image, i2v_elevation, i2v_center_scale, i2v_d_phi, i2v_d_theta, i2v_d_r, i2v_steps, i2v_seed],
+                            outputs=[i2v_traj_video,i2v_output_video],
+                            fn = image2video.run_gradio,
+                            cache_examples='lazy',
+                )
+
+            # image2video.run_gradio(i2v_input_image='test/images/boy.png', i2v_elevation='10', i2v_d_phi='0 40', i2v_d_theta='0 0', i2v_d_r='0 0', i2v_center_scale=1, i2v_steps=50, i2v_seed=123)
+            i2v_end_btn.click(inputs=[i2v_input_image, i2v_elevation, i2v_center_scale, i2v_d_phi, i2v_d_theta, i2v_d_r, i2v_steps, i2v_seed],
+                            outputs=[i2v_traj_video,i2v_output_video],
+                            fn = image2video.run_gradio
+            )
+
+    return viewcrafter_iface
+
+
+viewcrafter_iface = viewcrafter_demo(opts)
+viewcrafter_iface.queue(max_size=10)
+viewcrafter_iface.launch()
+# viewcrafter_iface.launch(server_name='127.0.0.1', server_port=80, max_threads=1,debug=False)
+

configs/inference_pvd_1024_local.yaml

@@ -0,0 +1,113 @@
+model:
+  pretrained_checkpoint: /apdcephfs_cq10/share_1290939/vg_share/vip3d_share/DC_1024/model.ckpt
+  base_learning_rate: 1.0e-05
+  scale_lr: False
+  target: lvdm.models.ddpm3d.VIPLatentDiffusion
+  params:
+    rescale_betas_zero_snr: True
+    parameterization: "v"
+    linear_start: 0.00085
+    linear_end: 0.012
+    num_timesteps_cond: 1
+    log_every_t: 200
+    timesteps: 1000
+    first_stage_key: video
+    cond_stage_key: caption
+    cond_stage_trainable: False
+    image_proj_model_trainable: True
+    conditioning_key: hybrid
+    image_size: [72, 128]
+    channels: 4
+    scale_by_std: False
+    scale_factor: 0.18215
+    use_ema: False
+    uncond_prob: 0.05
+    uncond_type: 'empty_seq'
+    rand_cond_frame: true
+    use_dynamic_rescale: true
+    base_scale: 0.3
+    fps_condition_type: 'fps'
+    perframe_ae: True
+    loop_video: Flase
+
+    unet_config:
+      target: lvdm.modules.networks.openaimodel3d.UNetModel
+      params:
+        in_channels: 8
+        out_channels: 4
+        model_channels: 320
+        attention_resolutions:
+        - 4
+        - 2
+        - 1
+        num_res_blocks: 2
+        channel_mult:
+        - 1
+        - 2
+        - 4
+        - 4
+        dropout: 0.1
+        num_head_channels: 64
+        transformer_depth: 1
+        context_dim: 1024
+        use_linear: true
+        use_checkpoint: True
+        temporal_conv: True
+        temporal_attention: True
+        temporal_selfatt_only: true
+        use_relative_position: false
+        use_causal_attention: False
+        temporal_length: 16
+        addition_attention: true
+        image_cross_attention: true
+        default_fs: 10
+        fs_condition: true
+
+    first_stage_config:
+      target: lvdm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        monitor: val/rec_loss
+        ddconfig:
+          double_z: True
+          z_channels: 4
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          - 4
+          num_res_blocks: 2
+          attn_resolutions: []
+          dropout: 0.0
+        lossconfig:
+          target: torch.nn.Identity
+
+    cond_stage_config:
+      target: lvdm.modules.encoders.condition.FrozenOpenCLIPEmbedder
+      params:
+        version: /apdcephfs_cq10/share_1290939/vg_share/vip3d_share/OpenCLIP-ViT-H-14-laion2B-s32B-b79K/blobs/9a78ef8e8c73fd0df621682e7a8e8eb36c6916cb3c16b291a082ecd52ab79cc4
+        freeze: true
+        layer: "penultimate"
+
+    img_cond_stage_config:
+      target: lvdm.modules.encoders.condition.FrozenOpenCLIPImageEmbedderV2
+      params:
+        version: /apdcephfs_cq10/share_1290939/vg_share/vip3d_share/OpenCLIP-ViT-H-14-laion2B-s32B-b79K/blobs/9a78ef8e8c73fd0df621682e7a8e8eb36c6916cb3c16b291a082ecd52ab79cc4
+        freeze: true
+    
+    image_proj_stage_config:
+      target: lvdm.modules.encoders.resampler.Resampler
+      params:
+        dim: 1024
+        depth: 4
+        dim_head: 64
+        heads: 12
+        num_queries: 16
+        embedding_dim: 1280
+        output_dim: 1024
+        ff_mult: 4
+        video_length: 16