Google Vision API - 标签检测多边形为空

Question

我在我的节点应用程序中使用了Image labels detection的Google vision API。 我正在尝试获取标签多边形，但它返回null。

const client = new visionClient.ImageAnnotatorClient();
client
    .labelDetection('./test.jpg')
    .then(results => {
        console.log(results[0]);
        const labels = results[0].labelAnnotations;

        console.log('Labels:');
        labels.forEach(label => console.log(label.description));
    })
    .catch(err => {
        console.error('ERROR:', err);
    });

结果：

[ { locations: [],
       properties: [],
       mid: '/m/0j272k5',
       locale: '',
       description: 'eyewear',
       score: 0.9805809855461121,
       confidence: 0,
       topicality: 0.9805809855461121,
       boundingPoly: null }]

boundingPoly：null