Instance level scene recognition with a vision language model
Systems and methods for image understanding can include one or more object recognition systems and one or more vision language models to generate an augmented language output that can be both scene-aware and object-aware. The systems and methods can process an input image with an object recognition model to generate an object recognition output descriptive of identification details for an object depicted in the input image. The systems and methods can include processing the input image with a vision language model to generate a language output descriptive of a predicted scene description. The object recognition output can then be utilized to augment the language output to generate an augmented language output that includes the scene understanding of the language output with the specificity of the object recognition output.
1. A computer-implemented method, the method comprising:
obtaining, by a computing system comprising one or more processors, image data and text data, wherein the image data comprises an input image, and wherein the text data comprises a query associated with the input image;
processing, by the computing system, the input image with an object recognition model to generate a fine-grained object recognition output, wherein the fine-grained object recognition output is descriptive of identification details for an object depicted in the input image;
processing, by the computing system, the input image and the text data with a vision language model to generate a language output, wherein the language output comprises a set of predicted words predicted to be responsive to the query and based on the input image, wherein the set of predicted words comprise a coarse-grained term descriptive of predicted identification of the object depicted in the input image; and
processing, by the computing system, the fine-grained object recognition output and the language output to generate an augmented language output, wherein the augmented language output comprises the set of predicted words with the coarse-grained term replaced with the fine-grained object recognition output.
2. The method of claim 1 , further comprising:
processing, by the computing system, the augmented language output with a search engine to determine a plurality of search results; and
providing, by the computing system, at least a subset of the plurality of search results as an output.
3. The method of claim 1 , further comprising:
processing, by the computing system, the augmented language output with a generative model to generate a model-generated response, wherein the model-generated response is responsive to a prompt in the augmented language output; and
providing, by the computing system, the model-generated response as an output.
4. The method of claim 3 , wherein the model-generated response comprises step-by-step instructions.
5. The method of claim 3 , wherein the model-generated response comprises multimodal data, wherein the multimodal data comprises one or more text strings and one or more images.
6. The method of claim 5 , wherein the one or more images are generated with a text-to-image generation model, wherein the one or more images are generated by processing the one or more text strings with the text-to-image generation model.
7. The method of claim 1 , wherein the language output is generated in a dialogue message format, and wherein the language output is a dialogue response to the query.
8. The system of claim 1 , wherein the vision language model comprises one or more image encoders to encode the input image, one or more text encoders to encode the text data, and one or more decoders to generate the language output.
9. The method of claim 1 , wherein object recognition model processing and vision language model processing are performed in parallel.
10. The method of claim 1 , wherein the language output is generated separately from the fine-grained object recognition output, and wherein the language output is generated without influence from the fine-grained object recognition output.
11. A computing system, the system comprising:
one or more processors; and
one or more non-transitory computer-readable media that collectively store instructions that, when executed by the one or more processors, cause the computing system to perform operations, the operations comprising:
obtaining image data and text data, wherein the image data comprises an input image, and wherein the text data comprises a question associated with the input image;
processing the input image to generate an object recognition output, wherein the object recognition output is descriptive of identification details for an object depicted in the input image, wherein processing the input image to generate the object recognition output comprises:
detecting the object in the input image;
generating an object embedding;
determining an image cluster associated with the object embedding; and
processing resource information associated with other embeddings in the image cluster to determine identification details for the object;
processing the input image with a vision language model to generate a language output, wherein the language output comprises a set of words predicted to be responsive to the text data and descriptive of features of the input image, wherein the set of words comprise a term descriptive of predicted identification of the object depicted in the input image; and
processing the object recognition output and the language output with the vision language model to generate an augmented language output, wherein the augmented language output comprises the set of words with the term replaced with the object recognition output.
12. The system of claim 11 , wherein generating the object embedding comprises:
generating a bounding box associated with a position of the object within the input image;
generating an image segment based on the bounding box; and
processing the image segment with an embedding model to generate the object embedding.
13. The system of claim 11 , wherein processing the object recognition output and the language output to generate the augmented language output comprises:
processing the language output to determine a plurality of text tokens associated with features in the input image;
determining a particular token of the plurality of text tokens is associated with the object; and
replacing, by the computing system, the particular token with the object recognition output.
14. The system of claim 13 , wherein determining, by the computing system, the particular token of the plurality of text tokens is associated with the object comprises:
processing, by the computing system, the fine-grained object recognition output with an embedding model to generate an instance-level embedding;
processing, by the computing system, the plurality of text tokens with the embedding model to generate a plurality of token embeddings; and
determining, by the computing system, the instance-level embedding is associated with a particular embedding associated with the particular token.
15. The system of claim 11 , wherein the object comprises an art piece, and wherein the object recognition output comprises a specific name for the art piece.
16. The system of claim 11 , wherein determining the image cluster associated with the object embedding comprises: searching an embedding space to determine a plurality of neighbor embeddings.
17. One or more non-transitory computer-readable media that collectively store instructions that, when executed by one or more computing devices, cause the one or more computing devices to perform operations, the operations comprising:
obtaining image data and text data, wherein the image data comprises an input image, and wherein the text data comprises a query associated with the input image;
processing the input image with an object recognition model to generate a fine-grained object recognition output, wherein the fine-grained object recognition output is descriptive of identification details for an object depicted in the input image;
processing the input image and the text data with a vision language model to generate a language output, wherein the language output comprises a set of predicted words predicted to be responsive to the query and based on the input image, wherein the set of predicted words comprise a coarse-grained term descriptive of predicted identification of the object depicted in the input image; and
processing the fine-grained object recognition output and the language output to generate an augmented language output, wherein the augmented language output comprises the set of predicted words with the coarse-grained term replaced with the fine-grained object recognition output.
18. The one or more non-transitory computer-readable media of claim 17 , wherein the coarse-grained term comprises an object type, and wherein the fine-grained object recognition output comprises a specific product label.
19. The one or more non-transitory computer-readable media of claim 18 , wherein the specific product label comprises a model name.
20. The one or more non-transitory computer-readable media of claim 17 , wherein the coarse-grained term comprises an object type, and wherein the fine-grained object recognition output comprises a name of a specific person.