CamScanner Translate Image#

Overview#

CamScanner provides image translation: detect text in images, translate to the target language, and render translated text preserving the original layout. The workflow is a 3-step pipeline: upload the image, translate it, then download the result.

When to Use#

• User wants to translate text in an image to another language
• User has a photo, screenshot, or scanned document with foreign text
• User wants to understand text in an image that is in a different language

Privacy & Data#

Important: Privacy & Data Flow Notice

• Third-party service: This skill sends your files to CamScanner's official servers ( ai-tools.camscanner.com) for processing.
• Data retention: CamScanner servers process your files in real-time. Files are not permanently stored on the server.
• Local files: Output files are saved to your local filesystem at the path you specify.

API Reference#

Base URL: https://ai-tools.camscanner.com

Target Language ( to parameter)#

The to parameter specifies the target language code. Determine it from the user's intent.

Common language mapping (use this first):

User says (English)User says (Chinese)to codeEnglish英语/英文enChinese Simplified简体中文/中文zh-HansChinese Traditional繁体中文zh-HantCantonese粤语yueJapanese日语/日文jaKorean韩语/韩文koFrench法语/法文frGerman德语/德文deSpanish西班牙语esPortuguese葡萄牙语ptRussian俄语/俄文ruArabic阿拉伯语arItalian意大利语itDutch荷兰语nlThai泰语/泰文thVietnamese越南语viIndonesian印尼语idMalay马来语msTurkish土耳其语trPolish波兰语plSwedish瑞典语svDanish丹麦语daHindi印地语hiBengali/Bangla孟加拉语bnUkrainian乌克兰语ukCzech捷克语csGreek希腊语elHebrew希伯来语heFinnish芬兰语fiHungarian匈牙利语hu

If the target language is NOT in the table above, look up the language code dynamically:

bash

LANG_DATA=$(curl -sS "https://open.camscanner.com/sync/get_languages")
# Parse LANG_DATA to find the matching language code by name or chineseName
# Example: find code for "Romanian"
TO=$(echo "$LANG_DATA" | jq -r '.data | to_entries[] | select(.value.name == "Romanian" or .value.chineseName == "罗马尼亚语") | .key')

The API returns all supported languages with name (English), nativeName, and chineseName (Chinese) fields.

Step 1: Upload Image#

bash

BASE="https://ai-tools.camscanner.com"

IN_FILE_ID=$(curl -sS -X POST "$BASE/v1/tools/upload_file/execute" \
  -H "Content-Type: application/octet-stream" \
  --data-binary "@/path/to/image.png" | jq -r '.tool_result.data.file_id')

Response:

json

{
  "code": 200,
  "tool": "upload_file",
  "tool_result": {
    "success": true,
    "data": {
      "file_id": "file_1741857600_ab12cd34ef56",
      "size": 24576
    }
  }
}

Step 2: Translate Image#

bash

OUT_FILE_ID=$(curl -sS -X POST "$BASE/v1/tools/translate_image/execute" \
  -H "Content-Type: application/json" \
  -d "{\"file_id\":\"$IN_FILE_ID\",\"to\":\"$TO\"}" \
  | jq -r '.tool_result.data.file_id')

Replace $TO with the target language code (e.g., en, ja, zh-Hans).

Response:

json

{
  "code": 200,
  "tool": "translate_image",
  "tool_result": {
    "success": true,
    "data": {
      "file_id": "file_1741857701_9988aabbccdd",
      "target_type": ""
    },
    "metadata": {
      "engine": "imagetranslate"
    }
  }
}

Step 3: Download Result#

bash

curl -sS -X POST "$BASE/v1/tools/download_file/execute?response_mode=raw" \
  -H "Content-Type: application/json" \
  -d "{\"file_id\":\"$OUT_FILE_ID\"}" \
  -o /path/to/translated.png

Critical: The response_mode=raw query parameter is required to get the binary file. Without it, the response is JSON.

Quick Reference: Complete Pipeline#

bash

BASE="https://ai-tools.camscanner.com"
INPUT_IMAGE="/path/to/image.png"
OUTPUT_FILE="/path/to/translated.png"
TO="en"   # target language code — see mapping table above

# Upload
IN_FILE_ID=$(curl -sS -X POST "$BASE/v1/tools/upload_file/execute" \
  -H "Content-Type: application/octet-stream" \
  --data-binary "@$INPUT_IMAGE" | jq -r '.tool_result.data.file_id')

# Translate
OUT_FILE_ID=$(curl -sS -X POST "$BASE/v1/tools/translate_image/execute" \
  -H "Content-Type: application/json" \
  -d "{\"file_id\":\"$IN_FILE_ID\",\"to\":\"$TO\"}" \
  | jq -r '.tool_result.data.file_id')

# Download
curl -sS -X POST "$BASE/v1/tools/download_file/execute?response_mode=raw" \
  -H "Content-Type: application/json" \
  -d "{\"file_id\":\"$OUT_FILE_ID\"}" \
  -o "$OUTPUT_FILE"

Language Code Lookup (for uncommon languages)#

If the user requests a language not in the common mapping table, look it up dynamically:

bash

TARGET_LANG="Romanian"   # or Chinese name like "罗马尼亚语"

LANG_DATA=$(curl -sS "https://open.camscanner.com/sync/get_languages")
TO=$(echo "$LANG_DATA" | jq -r --arg lang "$TARGET_LANG" \
  '.data | to_entries[] | select(.value.name == $lang or .value.chineseName == $lang or .value.nativeName == $lang) | .key')

if [ -z "$TO" ] || [ "$TO" = "null" ]; then
  echo "Unsupported language: $TARGET_LANG"; exit 1
fi

Common Mistakes#

MistakeFixForgetting response_mode=raw on downloadAlways append ?response_mode=raw to the download URLWrong Content-Type on uploadUpload uses application/octet-stream, not multipart/form-dataUsing GET instead of POSTAll three endpoints use POSTMissing to parameterAlways include target language code in the translate requestWrong language codeUse codes from the mapping table; for uncommon languages, use the lookup APIUsing "zh" instead of "zh-Hans"Chinese Simplified is zh-Hans, not zh; Traditional is zh-Hant

Error Handling#

Check each step before proceeding:

bash

# After upload
if [ -z "$IN_FILE_ID" ] || [ "$IN_FILE_ID" = "null" ]; then
  echo "Upload failed"; exit 1
fi

# After translate
if [ -z "$OUT_FILE_ID" ] || [ "$OUT_FILE_ID" = "null" ]; then
  echo "Translation failed"; exit 1
fi