Dynamic Scale Inference by Entropy Minimization
This work addresses scale variation in computer vision for tasks like semantic segmentation, offering an incremental improvement over existing dynamic models.
The paper tackles the problem of object recognition at varying scales by proposing an iterative optimization method with entropy minimization, which improves semantic segmentation accuracy and generalizes better to extreme scale variations compared to feedforward approaches.
Given the variety of the visual world there is not one true scale for recognition: objects may appear at drastically different sizes across the visual field. Rather than enumerate variations across filter channels or pyramid levels, dynamic models locally predict scale and adapt receptive fields accordingly. The degree of variation and diversity of inputs makes this a difficult task. Existing methods either learn a feedforward predictor, which is not itself totally immune to the scale variation it is meant to counter, or select scales by a fixed algorithm, which cannot learn from the given task and data. We extend dynamic scale inference from feedforward prediction to iterative optimization for further adaptivity. We propose a novel entropy minimization objective for inference and optimize over task and structure parameters to tune the model to each input. Optimization during inference improves semantic segmentation accuracy and generalizes better to extreme scale variations that cause feedforward dynamic inference to falter.