Beyond attention heatmaps

So far, we haven't actually specified what kind of analysis we're interested in. We took advantage of the default setting, which is a visual attention analysis that produces a attention heatmap. This isn't the only type of prediction available through the EyeQuant API, though.

Here's how we would configure a attention and a visual clarity analysis for the same image, and also specify which outputs we're interested in:

curl \
  -X POST \
  -H "Authorization: Bearer $apikey" \
  -H "Content-Type: application/json"  \
  -d '{
        "input": {
          "type": "image",
          "content": "iVBORw0KGgoAAAANSUhE....FTkSuQmCC",
          "medium": "desktopWeb",
          "title": "Example"
        "predictions": {
          "attention": {
          	"outputs": [ "attentionMap", "perceptionMap" ]
          "clarity": {
          	"outputs": [ "score", "map" ]

After processing has completed, the JSON response for the analysis would then look like this:

  "id": "611457618c1d4283a830d10a9ad4f8ae",
  "status": "success",
  "outputs": {
    "attention": {
      "attentionMap": "",
      "perceptionMap": ""
    "clarity": {
      "score": 87,
      "map": ""

For more information on available predictions and outputs, see the Reference.