Skeletonmae: graph-based masked autoencoder for skeleton sequence pre-training
Skeleton sequence representation learning has shown great advantages for action
recognition due to its promising ability to model human joints and topology. However, the …
recognition due to its promising ability to model human joints and topology. However, the …
Learning fine-grained view-invariant representations from unpaired ego-exo videos via temporal alignment
The egocentric and exocentric viewpoints of a human activity look dramatically different, yet
invariant representations to link them are essential for many potential applications in …
invariant representations to link them are essential for many potential applications in …
Modeling video as stochastic processes for fine-grained video representation learning
A meaningful video is semantically coherent and changes smoothly. However, most existing
fine-grained video representation learning methods learn frame-wise features by aligning …
fine-grained video representation learning methods learn frame-wise features by aligning …
Weakly supervised video representation learning with unaligned text for sequential videos
Sequential video understanding, as an emerging video understanding task, has driven lots
of researchers' attention because of its goal-oriented nature. This paper studies weakly …
of researchers' attention because of its goal-oriented nature. This paper studies weakly …
Learning viewpoint-agnostic visual representations by recovering tokens in 3d space
Humans are remarkably flexible in understanding viewpoint changes due to visual cortex
supporting the perception of 3D structure. In contrast, most of the computer vision models …
supporting the perception of 3D structure. In contrast, most of the computer vision models …
Universal time-series representation learning: A survey
Time-series data exists in every corner of real-world systems and services, ranging from
satellites in the sky to wearable devices on human bodies. Learning representations by …
satellites in the sky to wearable devices on human bodies. Learning representations by …
Context consistency regularization for label sparsity in time series
Labels are typically sparse in real-world time series due to the high annotation cost.
Recently, consistency regularization techniques have been used to generate artificial labels …
Recently, consistency regularization techniques have been used to generate artificial labels …
Solution of wide and micro background bias in contrastive action representation learning
S Liu, Z Luo, Y Li, Y Wang, W Fu, W Ding - Engineering Applications of …, 2024 - Elsevier
In recent years, contrastive learning has made great progress in the field of computer vision,
which shows great potential in action representation learning. Current contrastive learning …
which shows great potential in action representation learning. Current contrastive learning …
Conditional Information Bottleneck Approach for Time Series Imputation
Time series imputation presents a significant challenge because it requires capturing the
underlying temporal dynamics from partially observed time series data. Among the recent …
underlying temporal dynamics from partially observed time series data. Among the recent …
Action Detection for Wildlife Monitoring with Camera Traps Based on Segmentation with Filtering of Tracklets (SWIFT) and Mask-Guided Action Recognition …
F Schindler, V Steinhage, STS van Beeck Calkoen… - Applied Sciences, 2024 - mdpi.com
Behavioral analysis of animals in the wild plays an important role for ecological research
and conservation and has been mostly performed by researchers. We introduce an action …
and conservation and has been mostly performed by researchers. We introduce an action …