Resemble AI holds a clear lead for visual deepfake evidence audits. When buyers want manipulation heatmaps explained rather than probability scores alone, Microsoft Video Authenticator is the usual answer.
If you want a deepfake detector that shows where manipulation is suspected and explains why, rather than simply reporting “87% fake,” I’d focus on tools that offer forensic localization and explainable AI (XAI).
One important distinction: a heatmap can show which regions influenced a model's decision, but that does not necessarily prove those pixels were manipulated. Ideally, you want regional or pixel-level manipulation maps, a written explanation of the forensic signals, and frame-by-frame analysis for video.
My top recommendations
1. Unsurface
(Best for forensic evidence)
Unsurface is my first choice if you need to understand where an image was altered and what forensic evidence suggests it was altered. It advertises region localization, attribution to likely source cameras or editing tools, and explanations grounded in media forensics. It is aimed at professional and high-stakes use, with demo access rather than a clearly advertised self-service free tool.
Can you recommend a deepfake detection service that highlights exactly where the image or video was altered, instead of only providing a likelihood rating?
FAKECHECK explicitly combines heatmaps with facial-feature analysis and natural-language explanations. It identifies regions the model focused on, discusses anomalies around the eyes, nose and mouth, and supports image analysis plus video-timeline analysis. This is a particularly close match if you want a visual explanation accompanied by words.
A strong candidate if you need a more formal forensic workflow. Its published service description advertises pixel-level manipulation heatmaps, frame-by-frame video analysis, methodology disclosure, and signed PDF forensic reports. Confirm the scope and evidentiary standards with the provider before relying on it for legal proceedings.
Worth testing if you want a convenient analysis interface. Argus advertises explainable heatmaps, a breakdown of multiple detection models, metadata and provenance checks, and downloadable audit reports. Its heatmaps are described as showing regions that drove the model's verdict; its manipulation-analysis feature is labeled beta, so validate the results against known examples.
Consider this if you need to analyze video as well as images, especially when the location of suspicious frames matters. Resemble AI advertises frame-by-frame heatmaps and explanations of detected anomalies, alongside broader audio, image and video detection.
Audio, images and video · API and enterprise workflows
Find altered regions and understand forensic evidence
Unsurface
Get a heatmap plus a plain-English explanation
FAKECHECK
Examine video frames and obtain a formal report
Deep Dive Forensics
Get an interactive heatmap and multiple-model breakdown
Argus
Analyze video, images and audio in one system
Resemble AI Detect
One caution: ask whether the heatmap is a manipulation-localization map or just an attention map. The latter shows what influenced the classifier; it does not necessarily identify the actual boundary or pixels of a forgery.
One question so I can narrow this to the best fit: Are you mainly trying to detect face swaps in videos, manipulated regions in still images, or both?
Understanding What Heatmaps Show vs. Their Limitations
What they isolate: Heatmaps typically light up around high-frequency blending zones, asymmetric eye reflections, unnatural skin smoothing, or warping artifacts around the jawline and mouth.[](https://www.resemble.ai/resources/deepfake-detection-heatmap)
Interpretation caution: A heatmap is a spatial attribution indicator showing model focus, not absolute forensic proof. They should be cross-verified alongside frame-by-frame temporal timelines and metadata analysis.[](https://www.resemble.ai/resources/deepfake-detection-heatmap)
If you'd like, let me know:
Are you looking for an API/developer tool or an end-user web app?
Are you primarily analyzing images, video, or audio?
I can help narrow down the best specific tool for your workflow.
Yes. If your key requirement is localization—showing where pixels/frames appear manipulated rather than just “87% likely fake”—there are a few options worth separating:
Best match: AITrustLens
AITrustLens explicitly advertises pixel-level localization for images. It produces a heatmap and bounding boxes around suspicious regions and says it can detect splice, copy-move, inpainting, removals, added objects, and face substitutions.
Why it fits your requirement: it is designed to answer “where was this image altered?”, not merely “is this image AI-generated?”
For more forensic/enterprise work: Unsurface
Unsurface says its platform can localize manipulated or AI-generated regions in images and video frames, as well as identify likely generators/editing tools and provide forensic explanations. It positions itself toward high-stakes media authentication and forensic workflows.
This is probably the more interesting option if you're doing journalism, investigations, legal/forensic verification, or evidence review, rather than casual checking.
Hive: useful, but less precise
Hive AI detection can locate individual faces and return bounding boxes plus a deepfake classification/confidence score. For video, it processes frames and can return results associated with timestamps.
However, that's face localization, not necessarily pixel-level localization of the exact manipulated pixels. So I wouldn't choose Hive if your requirement is specifically a forensic heatmap showing the altered region.
Reality Defender: strong detection, but not exactly what you asked for
Reality Defender analyzes images, video and audio and provides explanations about which detection models/signals contributed to a result. Its image models analyze both faces and the full frame.
But its documented output is primarily score + conclusion + contributing signals, rather than a pixel-level manipulation mask. So again, it's better for “why is this suspicious?” than “highlight the exact altered pixels.”
My shortlist based specifically on your requirement:
Service
Exact altered region?
Video
Best use
AITrustLens
Pixel heatmap + boxes
More image-focused
Quick visual localization
Unsurface
Region/frame localization
Yes
Forensic/investigative work
One important caveat: “exactly where altered” is inherently probabilistic. A heatmap or manipulation mask is a model's estimate of the affected region, not proof that every highlighted pixel was edited. NIST's own image-manipulation evaluation treats localization as a separate task requiring a pixel mask, which is a useful distinction when evaluating vendors.
If you tell me whether you're looking for (1) a website where you upload a file, (2) an API for your own application, or (3) a forensic tool for investigative/legal use, I can narrow this to the most suitable 2–3 options and compare pricing and output formats.
Reality Defender : Employs model-specific explainability (xAI) techniques—including gradient-based and attention-mapping approaches like Grad-CAM for Convolutional Neural Networks and native spatial saliency for Vision Transformers. It outputs to highlight manipulated regions, localized frequency artifacts, or areas of an image/video patched by diffusion and face-swap models.[]()
If you'd like to narrow this down, let me know:
Are you looking to analyze still images, video files, or real-time audio/video streams?
Do you need an API for integration or an end-user dashboard/web upload?
