Search Results for author:
Found 35 papers, 26 papers with code
Order Matters: On Parameter-Efficient Image-to-Video Probing for Recognizing Nearly Symmetric Actions
and parameter-heavier PEFT-based transfer methods by 5-15%.
Mitigating Label Noise using Prompt-Based Hyperbolic Meta-Learning in Open-Set Domain Generalization
Open-Set Domain Generalization (OSDG) is a challenging task requiring models to accurately predict familiar categories while minimizing confidence for unknown categories to effectively reject them in unseen domains.
Advancing Open-Set Domain Generalization Using Evidential Bi-Level Hardest Domain Scheduler
In this paper, we observe that an adaptive domain scheduler benefits more in OSDG compared with prefixed sequential and random domain schedulers.
Towards Synthetic Data Generation for Improved Pain Recognition in Videos under Patient Constraints
Utilizing advanced facial capture techniques, and leveraging public datasets like CelebV-HQ and FFHQ-UV for demographic diversity, our new synthetic dataset significantly enhances model training while ensuring privacy by anonymizing identities through facial replacements.
Probing Fine-Grained Action Understanding and Cross-View Generalization of Foundation Models
This is the first systematic study of different foundation models and specific design choices for human activity recognition from unknown views, conducted with the goal to provide guidance for backbone- and temporal- fusion scheme selection.
Referring Atomic Video Action Recognition
Since these existing methods underperform on RAVAR, we introduce RefAtomNet -- a novel cross-stream attention-driven method specialized for the unique challenges of RAVAR: the need to interpret a textual referring expression for the targeted individual, utilize this reference to guide the spatial localization and harvest the prediction of the atomic actions for the referring person.
Skeleton-Based Human Action Recognition with Noisy Labels
In this study, we bridge this gap by implementing a framework that augments well-established skeleton-based human action recognition methods with label-denoising strategies from various research areas to serve as the initial benchmark.
Navigating Open Set Scenarios for Skeleton-based Action Recognition
In real-world scenarios, human actions often fall outside the distribution of training data, making it crucial for models to recognize known actions and reject unknown ones.
Quantized Distillation: Optimizing Driver Activity Recognition Models for Resource-Constrained Environments
The framework enhances 3D MobileNet, a neural architecture optimized for speed in video classification, by incorporating knowledge distillation and model quantization to balance model accuracy and computational efficiency.
Elevating Skeleton-Based Action Recognition with Efficient Multi-Modality Self-Supervision
These works overlooked the differences in performance among modalities, which led to the propagation of erroneous knowledge between modalities while only three fundamental modalities, i. e., joints, bones, and motions are used, hence no additional modalities are explored.
Exploring Self-supervised Skeleton-based Action Recognition in Occluded Environments
IosPSTL combines a cluster-agnostic KNN imputer with an Occluded Partial Spatio-Temporal Learning (OPSTL) strategy.
On Transferability of Driver Observation Models from Simulated to Real Environments in Autonomous Cars
For driver observation frameworks, clean datasets collected in controlled simulated environments often serve as the initial training ground.
Line Graphics Digitization: A Step Towards Full Automation
The digitization of documents allows for wider accessibility and reproducibility.
Exploring Few-Shot Adaptation for Activity Recognition on Diverse Domains
In this work, we focus on Few-Shot Domain Adaptation for Activity Recognition (FSDA-AR), which leverages a very small amount of labeled target videos to achieve effective adaptation.
AdaptiveClick: Clicks-aware Transformer with Adaptive Focal Loss for Interactive Image Segmentation
Interactive Image Segmentation (IIS) has emerged as a promising technique for decreasing annotation time.
FishDreamer: Towards Fisheye Semantic Completion via Unified Image Outpainting and Segmentation
This paper raises the new task of Fisheye Semantic Completion (FSC), where dense texture, structure, and semantics of a fisheye image are inferred even beyond the sensor field-of-view (FoV).
Towards Activated Muscle Group Estimation in the Wild
To this intent, we provide the MuscleMap dataset featuring >15K video clips with 135 different activities and 20 labeled muscle groups.
ModSelect: Automatic Modality Selection for Synthetic-to-Real Domain Generalization
Modality selection is an important step when designing multimodal systems, especially in the case of cross-domain activity recognition as certain modalities are more robust to domain shift than others.
Multimodal Generation of Novel Action Appearances for Synthetic-to-Real Recognition of Activities of Daily Living
We tackle this challenge and introduce an activity domain generation framework which creates novel ADL appearances (novel domains) from different existing activity modalities (source domains) inferred from video training data.
Multi-modal Depression Estimation based on Sub-attentional Fusion
Failure to timely diagnose and effectively treat depression leads to over 280 million people suffering from this psychological disorder worldwide.
Is my Driver Observation Model Overconfident? Input-guided Calibration Networks for Reliable and Interpretable Confidence Estimates
In this work, we for the first time examine how well the confidence values of modern driver observation models indeed match the probability of the correct outcome and show that raw neural network-based approaches tend to significantly overestimate their prediction quality.
A Comparative Analysis of Decision-Level Fusion for Multimodal Driver Behaviour Understanding
Visual recognition inside the vehicle cabin leads to safer driving and more intuitive human-vehicle interaction but such systems face substantial obstacles as they need to capture different granularities of driver behaviour while dealing with highly limited body visibility and changing illumination.
Towards Robust Semantic Segmentation of Accident Scenes via Multi-Source Mixed Sampling and Meta-Learning
Autonomous vehicles utilize urban scene segmentation to understand the real world like a human and react accordingly.
Ranked #1 on
Semantic Segmentation
on DADA-seg
(using extra training data)
TransDARC: Transformer-based Driver Activity Recognition with Latent Space Feature Calibration
This module operates in the latent feature-space enriching and diversifying the training set at feature-level in order to improve generalization to novel data appearances, (e. g., sensor changes) and general feature quality.
TransKD: Transformer Knowledge Distillation for Efficient Semantic Segmentation
Furthermore, we introduce two optimization modules to enhance the patch embedding distillation from different perspectives: (1) Global-Local Context Mixer (GL-Mixer) extracts both global and local information of a representative embedding; (2) Embedding Assistant (EA) acts as an embedding method to seamlessly bridge teacher and student models with the teacher's number of channels.
Delving Deep into One-Shot Skeleton-based Action Recognition with Diverse Occlusions
Yet, the research of data-scarce recognition from skeleton sequences, such as one-shot action recognition, does not explicitly consider occlusions despite their everyday pervasiveness.
Ranked #1 on
Action Classification
on Toyota Smarthome dataset
(Accuracy metric)
Should I take a walk? Estimating Energy Expenditure from Video Data
To study this underresearched task, we introduce Vid2Burn -- an omni-source benchmark for estimating caloric expenditure from video data featuring both, high- and low-intensity activities for which we derive energy expenditure annotations based on models established in medical literature.
Affect-DML: Context-Aware One-Shot Recognition of Human Affect using Deep Metric Learning
Human affect recognition is a well-established research area with numerous applications, e. g., in psychological care, but existing methods assume that all emotions-of-interest are given a priori as annotated training examples.
Transfer beyond the Field of View: Dense Panoramic Semantic Segmentation via Unsupervised Domain Adaptation
We look at this problem from the perspective of domain adaptation and bring panoramic semantic segmentation to a setting, where labelled training data originates from a different distribution of conventional pinhole camera images.
Ranked #7 on
Semantic Segmentation
on DensePASS
(using extra training data)
DensePASS: Dense Panoramic Semantic Segmentation via Unsupervised Domain Adaptation with Attention-Augmented Context Exchange
First, we formalize the task of unsupervised domain adaptation for panoramic semantic segmentation, where a network trained on labelled examples from the source domain of pinhole camera data is deployed in a different target domain of panoramic images, for which no labels are available.
Let's Play for Action: Recognizing Activities of Daily Living by Learning from Life Simulation Video Games
Recognizing Activities of Daily Living (ADL) is a vital process for intelligent assistive robots, but collecting large annotated datasets requires time-consuming temporal labeling and raises privacy concerns, e. g., if the data is collected in a real household.
MASS: Multi-Attentional Semantic Segmentation of LiDAR Data for Dense Top-View Understanding
At the heart of all automated driving systems is the ability to sense the surroundings, e. g., through semantic segmentation of LiDAR sequences, which experienced a remarkable progress due to the release of large datasets such as SemanticKITTI and nuScenes-LidarSeg.
Uncertainty-sensitive Activity Recognition: a Reliability Benchmark and the CARING Models
While temperature scaling alone drastically improves the reliability of the confidence values, our CARING method consistently leads to the best uncertainty estimates in all benchmark settings.
Multi-Task Learning for Calorie Prediction on a Novel Large-Scale Recipe Dataset Enriched with Nutritional Information
Our experiments demonstrate clear benefits of multi-task learning for calorie estimation, surpassing the single-task calorie regression by 9. 9%.
Informed Democracy: Voting-based Novelty Detection for Action Recognition
While it is common in activity recognition to assume a closed-set setting, i. e. test samples are always of training categories, this assumption is impractical in a real-world scenario.
