conflu3nce ImageOptix - Image/Scene Understanding

The Challenges of Efficacious Scene Understanding
Adaptive Image/Scene Perception Balancing Definition and Ambiguity	Figure and Ground/Depth Perception Real-time/Real-world Constraints

Integrative Information Processing Across Multiple Image Gestalt Scales (parts, part of the whole, and the whole)

ImageOptix - Visual Transformation Tools

conflu3nce's "image 1st" approach leverages principles of Gestalt to advance our understanding of figure-ground relationships, interactions between image parts/images, and to improve visual attention by reducing pixel content, color and/or complexity. The "minimum pixel requirement" supports scene analysis, and object detection and identification both in humans and machines.

CYCLOPS Intelligence (CI). "Stitch and Peel" (SnP) image manipulation methods can be applied to individual images or to image combinations to optimize the volume of total pixel content to improve detection sensitivity and accuracy. The process temporarily hyphenates pixel volume and content complexity by approximating non-contiguous regions. The new image constructs facilitate detection of asymmetries, saliency differences, and subtle ROIs and anomalies.

Intentional Ambiguity: Parts-of-a-Whole

Among the questions we ask, is how much information is needed to identify a part or reconstruct a scene. The related question is how we can maximize our understanding of the parts to generalize knowledge?

This can be likened to how children learn, to how they are able to take diverse representations of birds during early childhood and be able to see birds in all their inter-/intra-variability in color, shape, size and presentation format from cartoon characters to an eagle soaring in the sky - each in their complexity are simply, "BOORDS."

In our research, we seek to understand how spatially separated image parts in content-rich, stable and multi-stable image sets may be able to provide a unique opportunity to develop "logic leaping" training sets for holistically analyzing images, contexualizing long-range dependencies, while also considering an image's parts - a true image Gestalt.

The use of stitched images andthe role of Gestalt visual cues can be likened to distractors and attractors we humans must contend with when analyzing scenes. Image sets can contain flanking content which can be viewed as adversarial, or utilize spatially-separated image sections allowing deep learning networks to resolve the ambiguities, reconstruct the image scene and essentially solve what amounts to an image puzzle.

Patents/Patent Pending
10,582189 || 11,158,060 || 11,176,675 || 11,328,822 || 11,298,062 || 11,347,831 || 17/890,289