Search papers, labs, and topics across Lattice.
Addressing the fragmented embedding spaces caused by severe anatomical and ecological divergence in cross-species animal re-identification, the authors introduce Semantic Consistency Learning (SCL) to construct stable, unified identity representations. The framework decouples foreground animal representations from background habitat features using region-aware spectral normalization while enforcing shared relational structures across disparate taxonomies via dynamic neighborhood modeling. Across 11 benchmark animal datasets, SCL significantly outperforms prior methods under zero-shot and cross-species evaluation protocols, maintaining high retrieval precision on previously unseen species.
Animal visual re-identification no longer requires shared anatomical priors: decoupling environmental style from dynamic cross-species relational neighborhoods enables unified representation learning across radically different morphologies.
Generalizable animal Re-Identification (ReID) aims to recognize individual animals across species with diverse morphologies and ecological contexts. Unlike person ReID, where different domains share similar body structures, animal species often exhibit drastically different anatomical structures and visual patterns, making it difficult to establish shared visual correspondences. As a result, representations learned across species tend to form fragmented embedding spaces, which severely limits cross-species generalization. To address this challenge, we propose Semantic Consistency Learning (SCL), a framework designed to learn representations that remain stable across appearance variations while preserving semantic structures shared across species. SCL consists of two complementary components. Foreground-Background Decoupled Spectral Normalization (FDSNorm) stabilizes feature statistics by suppressing environment-induced style variations in a region-aware manner, while Cross-species Neighborhood Modeling (CNM) captures transferable relational structures across species through dynamic feature neighborhoods. Extensive experiments on 11 public animal ReID datasets demonstrate that SCL consistently outperforms state-of-the-art methods under multiple cross-species evaluation protocols and generalizes effectively to previously unseen species and ecological domains. Code is available at https://github.com/Kemalau/ECCV-26-SCL.
