I agree with Tilman. If there's a more modern package/model you'd want to use whether server based or commandline, it is fairly straightforward to add a new parser to handle your needs.
On Fri, Jul 26, 2024 at 11:50 PM Tilman Hausherr <[email protected]> wrote: > I don't think so, the closest we have is DL4J but I suspect that's > different than what you want. > > https://github.com/apache/tika/pull/165 > > Tilman > > On 26.07.2024 18:57, Kiran Bachu via user wrote: > > Hi All, > > We are trying to replace Google vision API with Tika. > > However, I cannot find any reference on whether Tika/Tessaract can > > support detecting labels/topics in an image. > > Google vision API detect labels - > > https://cloud.google.com/vision/docs/labels > > > > Appreciate any info. > > > > Thanks > > Kiran > > > > > >
