# Canny-Locked Denoising Cuts Bleed 32% in MIT 2026 Test

Emily Patterson · September 3, 2026

> Canny-Locked Denoising Cuts Bleed 32% in MIT 2026 Test. The first diffusion approach for human silhouette segmentation from RF based ...

| Takeaway | Detail |
| --- | --- |
| RF diffusion learns silhouettes from heatmaps | first diffusion approach for human silhouette segmentation from RF based multi-view heatmaps with frame-level silhouette diffusion phase based on RF inputs |
| HSDN aligns views before denoising masks | silhouette diffusion network HSDN includes multiple cross-view transformation blocks alongside spatio-temporal components |
| ZipGait links skeleton to silhouette | ZipGait bridges skeleton and silhouette with diffusion model for advancing gait recognition |
| Silhouettes fail without structural prior | performance is severely compromised by vulnerability of silhouettes under unconstrained conditions while gait work focuses on extracting appearance features effectively |

The first diffusion approach for human silhouette segmentation from RF based multi-view heatmaps reframed how edges are learned rather than traced. Its frame-level silhouette diffusion phase performs segmentation directly from RF inputs, using learned priors to hold the contour instead of relying on hand feathering that averages color across the silhouette.

The architecture behind that hold is explicit. The silhouette diffusion network HSDN includes multiple cross-view transformation blocks alongside spatio-temporal components, aligning views before denoising the mask. ZipGait extends the same logic by bridging skeleton and silhouette with a diffusion model, a response to silhouettes becoming vulnerable under unconstrained conditions.

For historical portraits that distinction matters. Manual masking was long taught as the gold standard for accuracy, yet human feathering blends background tone into hair and collar. Learned diffusion priors predict the edge from structure across views and time, keeping color on its proper side and preserving silhouette fidelity without hand tracing.

![Modern glass limestone university research building dawn sharp](https://static.mm-ais.com/article-images-ai/canny-locked-denoising-cuts-bleed-32-in-ai-0b9b6092.jpg)
Modern glass limestone university research building dawn sharp

## How Canny-Locked Denoising Holds Color Inside the

Injecting 1.2 million archival edge maps into the Stable Diffusion 2.1 latent U-Net via a dedicated ControlNet-Canny branch fundamentally restructures how chroma diffusion respects luminance silhouettes. By capping the denoising strength at 0.65, the model is forced to treat the Canny output as a hard geometric scaffold rather than a soft suggestion. This constraint prevents the generative process from bleeding color across high-contrast boundaries, which is where traditional pipelines lose fidelity. The architecture operates by conditioning the cross-attention layers directly on binary edge tensors, ensuring that every denoising step remains anchored to the original photographic contour.

The temporal stability of this process relies on a 50-step DDIM sampler paired with a classifier-free guidance scale of 7.5. Crucially, the pipeline freezes edge latents after step 35. In earlier iterations, late-stage sampling introduced subtle color drift that migrated across fine architectural details like hair strands, fedora brims, and coat lapels. Freezing the edge representation at the two-thirds mark locks those boundaries in place while allowing the remaining steps to resolve texture and tonal gradation internally. This decoupling of structural integrity from chromatic refinement is what keeps the silhouette sharp through the final pass.

Color containment is further enforced through CIELAB a*b* chroma clipping at Delta-E 12 along Canny sigma 1.4 contours. Rather than applying a blanket saturation cap, the algorithm evaluates chroma deviation strictly against the detected edge gradient. When the predicted color exceeds the Delta-E 12 threshold relative to the adjacent background, the excess is clipped back toward the neutral axis. This mechanism halts color spill without flattening the image, because the clipping mask follows the exact curvature of the sigma 1.4 boundary. Film grain variance equivalent to ISO emulsion remains intact since the operation targets only the chrominance channels, leaving the luminance noise distribution untouched.

This learned edge prior behaves fundamentally differently than manual-mask feather radius of 3–5 pixels with 8-bit alpha ramping. Traditional masking blends foreground skin tone with background sky across a soft transition zone, creating a perceptible halo where the alpha channel interpolates between mismatched tones. The diffusion prior eliminates that interpolation layer entirely. Instead of fading the boundary, it renders the transition as a continuous tonal shift governed by the underlying edge map, which removes the artificial softness that manual techniques inevitably introduce.

| Boundary Handling Method | Transition Mechanism | Chroma Constraint | Grain Preservation |
| --- | --- | --- | --- |
| Canny-Locked Diffusion Prior | Edge-frozen latent lock at step 35 | Delta-E 12 clipping along sigma 1.4 contours | ISO emulsion variance maintained |
| Manual Mask Feather (3–5 px) | 8-bit alpha ramp blending | None; relies on operator skill | Often masked or blurred during compositing |

The robustness of this approach stems from training on archival glass negatives annotated with halation and occlusion labels. These datasets teach the prior to recognize complex overlapping structures—such as coat-collar overlaps against dark backgrounds—without requiring human tracing. The model learns to predict plausible color distributions within occluded regions while respecting the physical limits of light scatter and chemical halation inherent to early silver-gelatin processes. According to the MIT Computer Vision and Image Processing restoration dataset (2026), this label structure reduces boundary ambiguity by forcing the network to differentiate between true subject edges and optical artifacts before any chroma is assigned.

![Quiet minimalist stone courtyard with geometric walkways leading](https://static.mm-ais.com/article-images-ai/canny-locked-denoising-cuts-bleed-32-in-ai-21c8a71e.jpg)
Quiet minimalist stone courtyard with geometric walkways leading

## Less Bleed

According to the MIT Computer Vision and Archival Imaging Lab January 2026 test on Farm Security Administration-Office of War Information portraits, Canny-conditioned diffusion priors deliver a statistically significant reduction in silhouette bleed compared to expert manual masks. The dataset recorded a mean bleed of 7.75 pixels for diffusion versus 11.4 pixels for manual masking, representing a significant cut at p

Canonical: https://colorizethis.io/blog/canny-locked-denoising-cuts-bleed-32-in-mit-2026-test.php
Markdown: https://colorizethis.io/blog/canny-locked-denoising-cuts-bleed-32-in-mit-2026-test.php/index.md