simoori.AI·IT 소식에서 신호만 골라 훑는 곳
피드로

Invideo가 GPT-6 Astra로 컬러 그레이딩 성공률을 3배 높였다

Invideo가 영상 편집 에이전트에 GPT-6 Astra를 도입해 컬러 작업 성공률을 세 배로 높였다.

OpenAI · 23일 오후 · 분야 동영상 편집 · 분야 AI 모델 · 사건종류 성능

원문: How invideo improves color grading 3x with GPT‑6 Astra

signal 포인트

핵심

  1. Invideo CEO 산켓 샤는 Astra가 프레임 단위로 정교하게 편집을 계획한다고 설명했다.
  2. Astra는 이전 모델보다 적은 추론 단계와 출력 토큰으로 복잡한 편집 작업을 수행한다.
  3. Invideo 편집자들은 Astra를 이용해 하루 만에 약 50개의 커스텀 효과를 만들어냈다.

요약은 자동으로 만들어집니다. 사실 확인이 필요하면 원문을 읽어주세요.

원문 보기 (영어)

아래는 출처에서 가져온 원문입니다.

Editing videos is one creative and technical decision after another. Editors have to shape the story, place sound effects, build transitions, adjust color, and more. Invideo⁠(opens in a new window) is an agentic video editor designed to handle that work while keeping the human editor in control.

With GPT‑6 Astra, invideo sees greater capability to plan and execute complex edits and place changes at the correct points in the timeline.

“How Astra can plan a particular edit on a frame-level accuracy is quite stunning.”

—Sanket Shah, CEO, invideo

Planning complex edits and staying on brief

The AI agent must translate the editor’s direction into a sequence of steps, choose the right tools, execute the work, and verify the result. Sanket says GPT‑6 Astra can plan these operations in fewer reasoning steps than the models it previously tested: “Astra is using far fewer reasoning sets and thus output tokens to complete complex work.”

Invideo also observed that Astra can carry out several instructions without losing the editor’s original objective as the task grows longer and more complex.

Choosing the right approach and adapting to change

GPT‑6 Astra is especially effective when it comes to color work, where the agent must choose from overlapping approaches such as correction, grading, regeneration, LUTs, and isolation. For example, changing a background while preserving a person’s skin tone requires the agent to isolate and track the person across frames before applying color elsewhere.

“Earlier, we saw very high failure rates for color-grading and color-correction tasks. With Astra, the success rate improved about three times,” said Sanket.

Creating custom effects that remain fully editable

GPT‑6 Astra also helps invideo turn descriptions and visual references into custom effects. The agent can code an effect tailored to the footage, place it on the timeline, and add controls that let the editor refine it. In one day, a few Invideo editors used the model to create about 50 effects.

With the invideo agent using GPT‑6 Astra, creative professionals retain control of the story, their taste, and the final cut without losing hours to the tedious planning and frame-by-frame work.

원문의 저작권은 출처에 있습니다.출처에서 원문 보기 새 창에서 열림