Is there any way to improve inference speed? #320

New Issue

claunia · 2026-01-29T21:46:49Z

claunia commented

2026-01-29 21:46:49 +00:00

Originally created by @Bruedream on GitHub (Mar 4, 2023).

I want to do real-time image processing. Is there any way to improve inference speed? Or can we reduce the size of the avatar from 512x512 to 256x256? I tried to modify several places in the code, changing 512 to 256, but I kept running errors. I don't know if it is related to the pre-training model? Which great god can guide you?

我想做图像的实时处理，有没有办法提升推理速度？或者可不可以把头像大小从512x512降低到256x256? 我尝试修改了代码中的几个地方，把512 改成了 256，但是一直运行错误，不知道是不是和预训练模型有关？哪位大神可以指导下？

Originally created by @Bruedream on GitHub (Mar 4, 2023). I want to do real-time image processing. Is there any way to improve inference speed? Or can we reduce the size of the avatar from 512x512 to 256x256? I tried to modify several places in the code, changing 512 to 256, but I kept running errors. I don't know if it is related to the pre-training model? Which great god can guide you? 我想做图像的实时处理，有没有办法提升推理速度？或者可不可以把头像大小从512x512降低到256x256? 我尝试修改了代码中的几个地方，把512 改成了 256，但是一直运行错误，不知道是不是和预训练模型有关？哪位大神可以指导下？

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: TencentARC/GFPGAN#320