To Fall Or Not To Fall: A Visual Approach to Physical Stability Prediction
This addresses the challenge of enabling AI systems to understand physical stability from observations, which is incremental as it applies an end-to-end method to a specific domain.
The paper tackles the problem of predicting physical stability of block towers from visual input, bypassing explicit simulation, and shows that their learning-based approach achieves results comparable to human judgments on synthetic data.
Understanding physical phenomena is a key competence that enables humans and animals to act and interact under uncertain perception in previously unseen environments containing novel object and their configurations. Developmental psychology has shown that such skills are acquired by infants from observations at a very early stage. In this paper, we contrast a more traditional approach of taking a model-based route with explicit 3D representations and physical simulation by an end-to-end approach that directly predicts stability and related quantities from appearance. We ask the question if and to what extent and quality such a skill can directly be acquired in a data-driven way bypassing the need for an explicit simulation. We present a learning-based approach based on simulated data that predicts stability of towers comprised of wooden blocks under different conditions and quantities related to the potential fall of the towers. The evaluation is carried out on synthetic data and compared to human judgments on the same stimuli.