Lackluster-Camera Translate

Yes, this is the third "wait, I've seen this node" moment in the pack's camera family - and again, it's the same node under two names. LacklusterQwenMultiangleCameraTranslateNode and LacklusterCameraTranslateNode are the same class, both displayed as "Lackluster-Camera Translate." One node, two registrations, identical inputs and outputs. If either name shows up in a shared workflow, the other works as a drop-in.

What it does

Takes a prompt and a target_language (中文 / 日本語 / 한국어 / English) and swaps the camera and shot phrases it knows for the target-language equivalents, using a fixed, hand-maintained glossary. Deterministic - no API, no LLM, no key. Everything outside its vocabulary passes through unchanged.

In:  <sks> low-angle shot wide shot
Out: <sks> 仰拍 远景

It's built to pair with the pack's own camera nodes: their prompt output uses exactly the phrases this glossary knows, so chaining Lackluster-Camera → this → your text encoder localizes an angle in one hop. If your model reads English camera terms fine, it's a no-op for you and you can ignore it.

Practical detail worth having

Because it's regex-based, it matches whole phrases, not substrings - "front view" inside a longer prompt still gets caught, but unrelated text is left alone. And because it's deterministic, the same input always produces the same output, which matters more than you'd think when you're iterating on a shot list: an LLM translator would happily give you four different translations of the same angle across four runs. This one won't.

Install

Pack-level only, no extra deps: ComfyUI Manager → "ComfyUI Lackluster Nodes" (requirements are just requests and numpy). And to close the loop on the naming: if you ever search this pack for "Translate" and see two identical entries, this is why - they're aliases, not upgrades.