即梦AI / 可灵AI / Veo / 海螺AI / 通义万相光影氛围镜头运镜
#光影氛围#镜头运镜
中文翻译
🕹️ 视频模型对比:I2V 多镜头
→ Wan 2.2
→ Veo 3
→ Seedance 1.0
→ Kling 2.1
→ Midjourney V1
→ Hailuo 02
我用相同的输入图像和多镜头提示词,在每个模型上快速对比了2-4次。
Wan 2.2 是最短的(我在 @replicate 上选择了3秒),但它在短短几秒内产生了几个不同的视角!
提示词如下:
原帖
🕹️ Video Model Comparison: I2V Multi-Shot
→ Wan 2.2
→ Veo 3
→ Seedance 1.0
→ Kling 2.1
→ Midjourney V1
→ Hailuo 02
I did a quick comparison with the same input image and multi-shot prompt on each model 2-4x.
Wan 2.2 was the shortest (I chose 3-sec on @replicate), but it produced several different perspectives in those few seconds!
Prompt below:
