7 papers
Liquid Neural Networks as a Drop-in Continuous-Time Deformation Field for Dynamic 3D Gaussian Splatting
Mingzhao Li, Arghya Pal, Guan Yuan Tan
Deformable 3D Gaussian Splatting (D-3DGS) re-constructs dynamic scenes from monocular video by deforming a canonical set of 3D Gaussians through a positional-encoded MLP of frame t…
Forgive or forget: Understanding the context of hate in audio retrieval systems
Arghya Pal, Sailaja Rajanala, Raphael C. -W. Phan +1
Handling toxic retrieval in text-to-audio systems is challenging due to contextual dependencies. Existing strategies (e.g., rephrasing, summarization) risk altering intent or omitt…
FORTE: FOL-guided Optimal Refinement for Text-audio rEtrieval
Arghya Pal, Sailaja Rajanala
Text-to-audio retrieval has made significant progress with shared embedding models such as CLAP and Pengi, yet they often struggle with fine-grained semantic alignment due to the i…
FLAG-4D: Flow-Guided Local-Global Dual-Deformation Model for 4D Reconstruction
Guan Yuan Tan, Ngoc Tuan Vu, Arghya Pal +4
We introduce FLAG-4D, a novel framework for generating novel views of dynamic scenes by reconstructing how 3D Gaussian primitives evolve through space and time. Existing methods ty…
Learning Energy-Based Generative Models via Potential Flow: A Variational Principle Approach to Probability Density Homotopy Matching
Junn Yong Loo, Michelle Adeline, Julia Kaiwen Lau +6
Energy-based models (EBMs) are a powerful class of probabilistic generative models due to their flexibility and interpretability. However, relationships between potential flows and…
Causal-Ex: Causal Graph-based Micro and Macro Expression Spotting
Pei-Sze Tan, Sailaja Rajanala, Arghya Pal +2
Detecting concealed emotions within apparently normal expressions is crucial for identifying potential mental health issues and facilitating timely support and intervention. The ta…