Search Results for author:
Found 5 papers, 2 papers with code
MeMSVD: Long-Range Temporal Structure Capturing Using Incremental SVD
Our scheme has two advantages: (a) it reduces complexity by more than an order of magnitude, and (b) it is amenable to an efficient implementation for the calculation of the memory bases in an incremental fashion which does not require the storage of the whole feature bank in memory.
VLLMs Provide Better Context for Emotion Understanding Through Common Sense Reasoning
In the first stage, we propose prompting VLLMs to generate descriptions in natural language of the subject's apparent emotion relative to the visual context.
Ranked #1 on
Emotion Recognition in Context
on EMOTIC
Multiscale Vision Transformers meet Bipartite Matching for efficient single-stage Action Localization
These methods build on adding a DETR head with learnable queries that after cross- and self-attention can be sent to corresponding MLPs for detecting a person's bounding box and action.
A Transfer Learning approach to Heatmap Regression for Action Unit intensity estimation
Action Units (AUs) are geometrically-based atomic facial muscle movements known to produce appearance changes at specific facial locations.
Attention-Aware Generative Adversarial Networks (ATA-GANs)
Using the attention maps produced by a Teacher- Network we are able to improve the quality of the generated images as well as perform weakly object localization on the generated images.
