# Adaptive Observation Control for Active Visual Recognition

arXiv: https://arxiv.org/abs/2610.19624 · HTML: https://mlsanity.com/paper/2610.19624/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 35 references that are in the map.
- First posted on arXiv 14 Oct 2026.

## Builds on (most cited first, 24 of 24)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Feature Pyramid Networks for Object Detection](https://mlsanity.com/paper/1612.03144/) (2016)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [RandAugment: Practical automated data augmentation with a reduced search space](https://mlsanity.com/paper/1909.13719/) (2019)
- [The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization](https://mlsanity.com/paper/2006.16241/) (2020)
- [Fine-Grained Visual Classification of Aircraft](https://mlsanity.com/paper/1306.5151/) (2013)
- [Spatial Transformer Networks](https://mlsanity.com/paper/1506.02025/) (2015)
- [Natural Adversarial Examples](https://mlsanity.com/paper/1907.07174/) (2019)
- [DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification](https://mlsanity.com/paper/2106.02034/) (2021)
- [BranchyNet: Fast Inference via Early Exiting from Deep Neural Networks](https://mlsanity.com/paper/1709.01686/) (2017)
- [Dynamic Neural Networks: A Survey](https://mlsanity.com/paper/2102.04906/) (2021)
- [AdaViT: Adaptive Tokens for Efficient Vision Transformer](https://mlsanity.com/paper/2112.07658/) (2021)
- [Recurrent Models of Visual Attention](https://mlsanity.com/paper/1406.6247/) (2014)
- [Depth-Adaptive Transformer](https://mlsanity.com/paper/1910.10073/) (2019)
- [Not All Images are Worth 16x16 Words: Dynamic Transformers for Efficient Image Recognition](https://mlsanity.com/paper/2105.15075/) (2021)
- [Part-based R-CNNs for Fine-grained Category Detection](https://mlsanity.com/paper/1407.3867/) (2014)
- [Glance and Focus: a Dynamic Approach to Reducing Spatial Redundancy in Image Classification](https://mlsanity.com/paper/2010.05300/) (2020)
- [FOVEA: Foveated Image Magnification for Autonomous Navigation](https://mlsanity.com/paper/2108.12102/) (2021)
- [FoveaTer: Foveated Transformer for Image Classification](https://mlsanity.com/paper/2105.14173/) (2021)
- [Emergent Properties of Foveated Perceptual Systems](https://mlsanity.com/paper/2006.07991/) (2020)
- [Emulating Human-like Adaptive Vision for Efficient and Flexible Machine Visual Perception](https://mlsanity.com/paper/2509.15333/) (2025)
- [Foveation in the Era of Deep Learning](https://mlsanity.com/paper/2312.01450/) (2023)
- [LLMind: Bio-inspired Training-free Adaptive Visual Representations for Vision-Language Models](https://mlsanity.com/paper/2603.14882/) (2026)

## Abstract (arXiv)

Visual recognition models usually process a fixed view under a fixed computation budget, although images often benefit from different scales and amounts of spatial evidence. Active vision can adapt by observing a coarse scene and then selecting informative fixations, but the public AdaptiveNN-DeiT-S pipeline still fixes both field of view and observation count. We extend this fixed-step recognizer with a controllable observation interface that adapts both the field of view and the number of fixations to the input. Specifically, the model predicts a per-fixation scale that sets its field of view, refines each selected crop with optional scale-gated foveation and a residual high-resolution detail-token branch, and uses label-grounded stop heads for input-dependent early termination. Experiments across fine-grained benchmarks, full ImageNet-1K, and ImageNet-A/R-derived classification splits show that the interface remains competitive with fixed-step references while exposing a controllable observation budget. The same checkpoint enables threshold-controlled accuracy-compute trade-offs without per-threshold retraining. On sequential ImageNet-1K, the learned stop rule outperforms fixed-step truncation of the same checkpoint, sustaining 81.9% top-1 using 2.45 of the four focus steps and exceeding three-step truncation at lower cost. On each benchmark, a variant combining scale control and detail-token fusion achieves the highest accuracy among the tested component-added non-exit variants. Our code is available at https://github.com/GZHU-DVL/LI-Dikai.