Google’s Bard recently received a fantastic update! Users can now upload images with prompts, thanks to the seamless integration of Google Lens. This new feature allows you to provide extra context to your queries or simply have some fun with the model.
The recent integration of Google Lens into Bard is seamlessly incorporated, making it appear as an inherent feature of the platform. When you upload images to Bard, you have the option to provide additional context to your prompts. This can involve enriching background information for the language model to process or seeking input directly from Bard.
For example, Google suggests uploading an image and asking Bard to generate a humorous caption. On a personal note, I’ve been having an enjoyable time challenging the model by inquiring about the image’s contents to evaluate its recognition capabilities. It’s also important to note that Bard’s new Google Lens integration is currently available in English only at the time of writing.
Uploading Images to Bard with Google Lens
- To get started, head over to bard.google.com on your web browser.
- Click the plus icon next to the text field, and select an image from your device.
- Add some context to the photo by entering a query or command.
Discovering Bard’s Responses
Once you’ve uploaded an image and paired it with a query, Bard will provide you with an educated reply. For instance, if you ask Bard to identify objects in the picture, Google Lens will diligently analyze the image and generate responses accordingly.
While it works hard to provide accurate information, some answers may occasionally be incorrect, especially in complex images. For instance, I asked it to identify the various objects in the picture shown below.
And this was the answer Bard came up with.
Understandably, Google Lens worked as hard as it could to identify a table with a charger, a picture of an atom, and a bunch of wires connected to a Starlink satellite dish. However, you can clearly tell that some of the answers it gave we’re wrong. The image is from the Starlink Hardware site, not the Starlink website. The atom picture is also not mounted on the wall but rather engraved on the Stalink router.
Refining Bard’s Responses
As Bard excels in conversations, you can ask the model to try again or refine its responses, especially when seeking funny captions for images or engaging in playful interactions. The integration with Google Lens enhances the overall Bard experience, making it an engaging language model experiment for users.
Additional Tips
The recent update not only includes Google Lens integration but also introduces new pins. By clicking the three-dot menu at the top, you can hit “Pin” to save conversations for easy recall later. This feature further enhances your interaction with Bard, making it convenient to revisit conversations and explore new prompts.
Conclusion
Google’s Bard continues to evolve and captivate users with its innovative features. The latest Google Lens integration adds a new dimension to the language model experiment, enabling users to upload images with prompts for context or amusement. As the company expands Bard, more exciting features are likely to emerge, enhancing the model’s potential and enriching the overall user experience.
