Animal Identification with Independent Foreground and Background Modeling
This work addresses animal identification for ecological monitoring, presenting an incremental improvement through independent modeling of foreground and background.
The paper tackles the problem of visual identification of individual animals by robustly exploiting background and foreground information, achieving a 22.3% and 8.8% reduction in relative error compared to baselines and doubling accuracy in cases of background drift.
We propose a method that robustly exploits background and foreground in visual identification of individual animals. Experiments show that their automatic separation, made easy with methods like Segment Anything, together with independent foreground and background-related modeling, improves results. The two predictions are combined in a principled way, thanks to novel Per-Instance Temperature Scaling that helps the classifier to deal with appearance ambiguities in training and to produce calibrated outputs in the inference phase. For identity prediction from the background, we propose novel spatial and temporal models. On two problems, the relative error w.r.t. the baseline was reduced by 22.3% and 8.8%, respectively. For cases where objects appear in new locations, an example of background drift, accuracy doubles.