TAPe+ML: A Compact Structured Representation for Multi-Task Computer Vision
TAPe+ML v3 reaches 84.7 COCO detection mAP50 with under 100,000 parameters via a structured Theory of Active Perception representation.
TAPe+ML v3 is a compact multi-task computer vision system that encodes relations among perceptual elements in a shared TAPe representation instead of raw pixel tensors, combining contour processing, localization, prototype classification, and a coordinator over submodels. Using fewer than 100,000 parameters, it scores 84.7 mAP50 on COCO detection, 80.7 mask mAP50 on COCO segmentation, and 89.9 percent top-1 on ImageNet-Real. Compactness and distribution-shift adaptation were additionally evaluated in an industrial pilot.
- Under 100,000 parameters across classification, detection, and segmentation tasks
- 84.7 mAP50 COCO detection; 80.7 mask mAP50 COCO segmentation
- Structured representation shifts modeling burden away from network parameters
Full article180 words · extracted from huggingface.co · click to collapse
We present TAPe+ML v3, a compact computer vision system based on TAPe (Theory of Active Perception), a structured representation that encodes relations among perceptual elements before recognition. Instead of operating directly on pixel tensors, the system uses a shared TAPe representation and a modular recognition architecture for image classification, object detection, and instance segmentation. TAPe+ML v3 combines background and contour processing, local object localization, prototype-based classification, and a coordinator for specialized submodels. Across the reported experiments, it uses fewer than 100,000 parameters. On COCO object detection, it obtains 84.7 mAP50 and 65.3 mAP50-95. On COCO instance segmentation, it obtains 80.7 mask mAP50 and 58.4 mask mAP50-95. In classification experiments, it reaches 92 percent validation accuracy on Imagenette under an identical-training comparison with a raw-pixel baseline, and 89.9 percent Top-1 accuracy on ImageNet-Real. We also evaluate compactness in video scene detection and adaptation under distribution shift in an industrial pilot. The results suggest that shifting part of the modeling burden from network parameters to a structured input representation can support compact multi-task vision systems with reduced data, memory, and compute requirements.
Text extracted automatically; images, tables and formatting may be missing. Original: https://huggingface.co/papers/2609.20869