Vivid is a simple Google Vision java client with much simple api and less dependencies.
RequestBuilder builder = new RequestBuilder()
.content(toBase64(new File("test/text.jpg")))
.addFeature(FeatureType.TEXT_DETECTION);
Request request = builder.build();
List<Request> requests = new ArrayList<>();
requests.add(request);
Result result = post("test text OCR", requests);
TextAnnotation ann = result.getResponses().get(0).getFullTextAnnotation().get(0);
assertNull(result.getErrors());
assertTrue(ann.getText().contains("GPS"));
assertTrue(ann.getText().contains("3.81730"));
assertTrue(ann.getText().contains("永泰饼家"));
dependecies:
compile 'com.alibaba:fastjson:1.+'
compile 'com.squareup.okhttp3:okhttp:3.+'
Refer to unit test for difference scenario.
The Version is sync with Google Vision version. E.g. vivid version 1.0 == Google Vision v1.