---
title: "Follow an object through a video (a mask video) — estimate"
method: POST
path: "/v1/ai/track-object/estimate"
tags: ["Object tracking"]
---

# Follow an object through a video (a mask video) — estimate

`POST /v1/ai/track-object/estimate`

## Request

- Base URL: `https://apihub.co.il`
- URL: `https://apihub.co.il/v1/ai/track-object/estimate`
- Auth: HTTP bearer

Price the request without executing it. Same body as `POST /v1/ai/track-object`.

## Request body

- object
  - `output` 'mask' | 'overlay': mask (the default) = a black-and-white video of the same length, white where the object is — for effects that follow it · overlay = the object highlighted over the video (a preview)
  - `prompt` string, required: What to follow, in a few words: "the bottle", "the red car" — several things with commas: "person, dog"
  - `video_url` string, uri, required: The video (up to 60 s)
  - `duration_seconds` number: The video's length if you know it (only for the estimate)
  - `detection_threshold` number: How sure the model must be to follow something (lower finds more, less precisely; try 0.3 when nothing is found)

## Response `200`

Estimate

- Estimate
  - `operation` string
  - `estimated_credits` number
  - `credits_to_reserve` number
  - `estimated_usage` object
  - `note` string, nullable
  - `balance` object
    - `available` number
  - `sufficient` boolean

---

[API](https://skmtc.dev/apihub/apis/apihub-one-api-for-chat-image-video-ai.md) · [All operations](https://skmtc.dev/apihub/apis/apihub-one-api-for-chat-image-video-ai/llms.txt) · [OpenAPI document](https://skmtc.dev/apihub/apis/apihub-one-api-for-chat-image-video-ai/revisions/a8caa42835ce?raw)