A Light-Weight Contrastive Approach for Aligning Human Pose Sequences
원문 초록
We present a simple unsupervised method for learning an encoder mapping short 3D pose sequences into embedding vectors suitable for sequence-to-sequence alignment by dynamic time warping. Training samples consist of temporal windows of frames containing 3D body points such as mocap markers or skeleton joints. A light-weight, 3-layer encoder is trained using a contrastive loss function that encourages embedding vectors of augmented sample pairs to have cosine similarity 1, and similarity 0 with all other samples in a minibatch. When multiple scripted training sequences are available, temporal alignments inferred from an initial round of training are harvested to extract additional, cross-performance match pairs for a second phase of training to refine the encoder. In addition to being simple, the proposed method is fast to train, making it easy to adapt to new data using different marker sets or skeletal joint layouts. Experimental results illustrate ease of use, transferability, and utility of the learned embeddings for comparing and analyzing human behavior sequences.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
ActAlign: Zero-Shot Fine-Grained Video Classification via Language-Guided Sequence Alignment
TimePoint: Accelerated Time Series Alignment via Self-Supervised Keypoint and Descriptor Learning
Semi-Supervised Image-Based Narrative Extraction: A Case Study with Historical Photographic Records
Automatic Differentiation-based Full Waveform Inversion with Flexible Workflows
Contrast Similarity-Aware Dual-Pathway Mamba for Multivariate Time Series Node Classification
Multiscale Dubuc: A New Similarity Measure for Time Series
토론
이 논문과 연결된 포럼 글이 아직 없습니다.