Technical Report: Masked Skeleton Sequence Modeling for Learning Larval Zebrafish Behavior Latent Embeddings
This work addresses behavior analysis for larval zebrafish researchers, but it appears incremental as it adapts existing masked modeling techniques to a new domain.
The authors tackled the problem of extracting latent embeddings from larval zebrafish behavior sequences by introducing a novel self-supervised learning method, resulting in a proposed Transformer-CNN architecture called SSTFormer that captures spatial-temporal correlations without providing concrete performance numbers.
In this report, we introduce a novel self-supervised learning method for extracting latent embeddings from behaviors of larval zebrafish. Drawing inspiration from Masked Modeling techniquesutilized in image processing with Masked Autoencoders (MAE) \cite{he2022masked} and in natural language processing with Generative Pre-trained Transformer (GPT) \cite{radford2018improving}, we treat behavior sequences as a blend of images and language. For the skeletal sequences of swimming zebrafish, we propose a pioneering Transformer-CNN architecture, the Sequence Spatial-Temporal Transformer (SSTFormer), designed to capture the inter-frame correlation of different joints. This correlation is particularly valuable, as it reflects the coordinated movement of various parts of the fish body across adjacent frames. To handle the high frame rate, we segment the skeleton sequence into distinct time slices, analogous to "words" in a sentence, and employ self-attention transformer layers to encode the consecutive frames within each slice, capturing the spatial correlation among different joints. Furthermore, we incorporate a CNN-based attention module to enhance the representations outputted by the transformer layers. Lastly, we introduce a temporal feature aggregation operation between time slices to improve the discrimination of similar behaviors.