Translations:Diffusion Models Are Real-Time Game Engines/56/zh: Difference between revisions

Latest revision as of 00:26, 9 September 2024

Information about message (contribute)

This message has no documentation. If you know where or how this message is used, you can help other translators by adding documentation to this message.

Message definition (Diffusion Models Are Real-Time Game Engines)

'''Video Quality.''' We use the auto-regressive setup described in Section [https://arxiv.org/html/2408.14837v1#S2 2], where we iteratively sample frames following the sequences of actions defined by the ground-truth trajectory, while conditioning the model on its own past predictions. When sampled auto-regressively, the predicted and ground-truth trajectories often diverge after a few steps, mostly due to the accumulation of small amounts of different movement velocities between frames in each trajectory. For that reason, per-frame PSNR and LPIPS values gradually decrease and increase respectively, as can be seen in Figure [https://arxiv.org/html/2408.14837v1#S5.F6 6]. The predicted trajectory is still similar to the actual game in terms of content and image quality, but per-frame metrics are limited in their ability to capture this (see Appendix [https://arxiv.org/html/2408.14837v1#A1.SS1 A.1] for samples of auto-regressively generated trajectories).

视频质量我们使用第2节中描述的自回归设置，按照真实轨迹所定义的动作序列对帧进行迭代采样，同时将模型自身的过往预测作为条件。自回归采样时，预测轨迹和真实轨迹常常在几步后发生偏离，这主要是由于不同轨迹的帧间积累了少量不同的运动速度。因此，如图6所示，每帧的PSNR和LPIPS值分别逐渐降低和增加。预测的轨迹在内容和图像质量方面仍与实际游戏相似，但每帧指标在捕捉这一点的能力上有限（自动回归生成的轨迹样本见附录A.1）。