Merge c5a6d87db7 into ae487cc653

Co-authored-by: Shiwei Zhang <134917139+Steven-SWZhang@users.noreply.github.com>

README.md

@@ -36,6 +36,7 @@ In this repository, we present **Wan2.1**, a comprehensive and open suite of vid
 
 ## Community Works
 If your work has improved **Wan2.1** and you would like more people to see it, please inform us.
+- [Video-As-Prompt](https://github.com/bytedance/Video-As-Prompt), the first unified semantic-controlled video generation model based on **Wan2.1-14B-I2V** with a Mixture-of-Transformers architecture and in-context controls (e.g., concept, style, motion, camera). Refer to the [project page](https://bytedance.github.io/Video-As-Prompt/) for more examples.
 - [LightX2V](https://github.com/ModelTC/LightX2V), a lightweight and efficient video generation framework that integrates **Wan2.1** and **Wan2.2**, supports multiple engineering acceleration techniques for fast inference, which can run on RTX 5090 and RTX 4060 (8GB VRAM).
 - [DriVerse](https://github.com/shalfun/DriVerse), an autonomous driving world model based on **Wan2.1-14B-I2V**, generates future driving videos conditioned on any scene frame and given trajectory. Refer to the [project page](https://github.com/shalfun/DriVerse/tree/main) for more examples.
 - [Training-Free-WAN-Editing](https://github.com/KyujinHan/Awesome-Training-Free-WAN2.1-Editing), built on **Wan2.1-T2V-1.3B**, allows training-free video editing with image-based training-free methods, such as [FlowEdit](https://arxiv.org/abs/2412.08629) and [FlowAlign](https://arxiv.org/abs/2505.23145).

wan/first_last_frame2video.py

@@ -13,6 +13,7 @@ import numpy as np
 import torch
 import torch.cuda.amp as amp
 import torch.distributed as dist
+import torchvision
 import torchvision.transforms.functional as TF
 from tqdm import tqdm
 
@@ -211,7 +212,12 @@ class WanFLF2V:
                 round(last_frame_size[1] * last_frame_resize_ratio),
             ]
             # 2. center crop
-            last_frame = TF.center_crop(last_frame, last_frame_size)
+            transform = torchvision.transforms.Compose([
+                torchvision.transforms.Resize((last_frame_size[0], last_frame_size[1])),
+                torchvision.transforms.CenterCrop((first_frame_size[0], first_frame_size[1]))
+            ])
+
+            last_frame = transform(last_frame)
 
         max_seq_len = ((F - 1) // self.vae_stride[0] + 1) * lat_h * lat_w // (
             self.patch_size[1] * self.patch_size[2])