CVAIJul 1

Learning When to Listen: Gated Affect Fusion for Human Motion Prediction

arXiv:2607.002961.1
Predicted impact top 97% in CV · last 90 daysOriginality Incremental advance
AI Analysis

This work provides empirical guidance for selective multimodal fusion in human motion forecasting, clarifying when and how facial affect should be used.

The paper investigates the utility of facial affect for human motion prediction in unconstrained videos, finding that naive fusion degrades performance while a gated mechanism (GAT) improves short-to-medium term forecasting (e.g., 30 frames), but affect provides bounded, horizon-dependent cues and long-term motion is governed by kinematics.

Human motion forecasting in unconstrained real-world videos remains challenging due to the ambiguity of future behaviors and the presence of noisy multimodal observations. While facial affect potentially provides complementary behavioral cues, its practical utility and mechanistic boundaries within motion forecasting frameworks remain poorly understood. In this work, we present a systematic study investigating the utility and temporal limitations of affect-conditioned forecasting in-the-wild. We establish a rigorous multimodal pipeline combining MediaPipe body pose trajectories with HSEmotion facial affect representations, and introduce the Gated Affect Transformer (GAT) to dynamically regulate cross-modal information flow. Through extensive multi-horizon evaluations under a strict subject-wise protocol, we demonstrate that naive early cross-modal concatenation consistently degrades forecasting accuracy relative to pose-only baselines. Conversely, our proposed gating mechanism stabilizes cross-modal integration by adaptively controlling the affective stream. Crucially, controlled counterfactual experiments using shuffled and randomized affect inputs reveal that the learned gate successfully suppresses unstructured cross-modal noise while remaining responsive to plausible affective signals. Furthermore, our empirical results indicate that facial affect features provide bounded, horizon-dependent predictive cues strictly within short-to-medium windows (e.g., 30 frames), whereas long-term trajectories remain predominantly governed by intrinsic kinematic continuity. Our findings provide empirical evidence that facial affect should be regarded as a complementary behavioral cue rather than a dominant driver of future motion, offering practical guidance for selective multimodal fusion in unconstrained human motion forecasting.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes