Repository navigation
Home
It is intended for investigators who have a directory of evidence—images, audio, PDFs, text files, and want to run AI/ML analysis.
This is done local without uploading that evidence to a cloud service.
“Transcribe every audio recording in this directory.”
“Find photographs matching ‘a person wearing a red jacket.’”
“Find other images from the same event as this photo.”
“Describe these images, then search their descriptions.”
“Find faces resembling the supplied query photographs.”
“Estimate the age range and gender presentation of detected faces.”
“Flag images that may be manipulated or AI-generated.”
“Summarize all PDFs in this evidence folder.”
Create or load an investigation case.
Select an analysis plugin from a Menu or promt the desired analysis to the Chat assistant.
Fill in inputs like a local evidence directory. fill in or select ML parameters like model to use,top-k.
Submit a job.
Inspect the results and job history.
Optionally chain several tools into a pipeline.
Special thanks to Brian Levine.
Open any issue you feel is imporant here.
Please refer to the CONTRIBUTING.md file for more information on how to contribute to this project.
Brian Levine brian at cs dot umass dot edu