Search Results for author:
Found 55 papers, 9 papers with code
Neural Batch Sampling with Reinforcement Learning for Semi-Supervised Anomaly Detection
In this work, our key hypothesis is that this change in loss values during training can be used as a feature to identify anomalous data.
Don't Look Twice: Faster Video Transformers with Run-Length Tokenization
Our method is content-aware, requiring no tuning for different datasets, and fast, incurring negligible overhead.
Zero-Shot Video Question Answering with Procedural Programs
We propose to answer zero-shot questions about videos by generating short procedural programs that derive a final answer from solving a sequence of visual subtasks.
Ranked #15 on
Zero-Shot Video Question Answer
on NExT-QA
Deformer: Dynamic Fusion Transformer for Robust Hand Pose Estimation
Accurately estimating 3D hand pose is crucial for understanding how humans interact with the world.
HERD: Continuous Human-to-Robot Evolution for Learning from Human Demonstration
The ability to learn from human demonstration endows robots with the ability to automate various tasks.
From Universal Humanoid Control to Automatic Physically Valid Character Creation
Second, we use a design-and-control framework to optimize a humanoid's physical attributes to find body designs that can better imitate the pre-specified human motion sequence(s).
Cost-Aware Evaluation and Model Scaling for LiDAR-Based 3D Object Detection
While progress has been encouraging, we observe an overlooked issue: it is not yet common practice to compare different 3D detectors under the same cost, e. g., inference latency.
Domain Adaptive Hand Keypoint and Pixel Localization in the Wild
We aim to improve the performance of regressing hand keypoints and segmenting pixel-level hand masks under new imaging conditions (e. g., outdoors) when we only have labeled images taken under very different conditions (e. g., indoors).
REvolveR: Continuous Evolutionary Models for Robot-to-robot Policy Transfer
We interpolate between the source robot and the target robot by finding a continuous evolutionary change of robot parameters.
V-MAO: Generative Modeling for Multi-Arm Manipulation of Articulated Objects
Manipulating articulated objects requires multiple robot arms in general.
Sequential Voting with Relational Box Fields for Active Object Detection
While our voting function is able to improve the bounding box of the active object, one round of voting is typically not enough to accurately localize the active object.
Adaptive Unbiased Teacher for Cross-Domain Object Detection
This enables the student model to capture domain-invariant features.
StereOBJ-1M: Large-scale Stereo Image Dataset for 6D Object Pose Estimation
We present a large-scale stereo RGB image object pose estimation dataset named the $\textbf{StereOBJ-1M}$ dataset.
KDFNet: Learning Keypoint Distance Field for 6D Object Pose Estimation
To address this problem, we propose a novel continuous representation called Keypoint Distance Field (KDF) for projected 2D keypoint locations.
No-Reference Image Quality Assessment via Transformers, Relative Ranking, and Self-Consistency
Specifically, we enforce self-consistency between the outputs of our quality assessment model for each image and its transformation (horizontally flipped) to utilize the rich self-supervisory information and reduce the uncertainty of the model.
Ranked #3 on
No-Reference Image Quality Assessment
on TID2013
Neighborhood-Aware Neural Architecture Search
To facilitate the application to gradient-based algorithms, we also propose a differentiable representation for the neighborhood of architectures.
RePOSE: Fast 6D Object Pose Refinement via Deep Texture Rendering
Furthermore, we utilize differentiable Levenberg-Marquardt (LM) optimization to refine a pose fast and accurately by minimizing the feature-metric error between the input and rendered image representations without the need of zooming in.
Ranked #5 on
6D Pose Estimation using RGB
on LineMOD
Visio-Temporal Attention for Multi-Camera Multi-Target Association
We propose a inter-tracklet (person to person) attention mechanism that learns a representation for a target tracklet while taking into account other tracklets across multiple views.
Wisdom of Committees: An Overlooked Approach To Faster and More Accurate Models
Committee-based models (ensembles or cascades) construct models by combining existing pre-trained ones.
Kinematics-Guided Reinforcement Learning for Object-Aware 3D Ego-Pose Estimation
We propose a method for incorporating object interaction and human body dynamics into the task of 3D ego-pose estimation using a head-mounted camera.
3D Human Motion Estimation via Motion Compression and Refinement
Experiments show that our method produces both smooth and accurate 3D human pose and motion estimates.
Ranked #16 on
3D Human Pose Estimation
on 3DPW
(Acceleration Error metric, using extra
training data)
Importance of Self-Consistency in Active Learning for Semantic Segmentation
We address the task of active learning in the context of semantic segmentation and show that self-consistency can be a powerful source of self-supervision to greatly improve the performance of a data-driven model with access to only a small amount of labeled data.
AttentionNAS: Spatiotemporal Attention Cell Search for Video Classification
The discovered attention cells can be seamlessly inserted into existing backbone networks, e. g., I3D or S3D, and improve video classification accuracy by more than 2% on both Kinetics-600 and MiT datasets.
Learning Shape Representations for Clothing Variations in Person Re-Identification
To tackle the re-ID problem in the context of clothing changes, we propose a novel representation learning model which is able to generate a body shape feature representation without being affected by clothing color or patterns.
Human Motion Trajectory Prediction: A Survey
With growing numbers of intelligent autonomous systems in human environments, the ability of such systems to perceive, understand and anticipate human behavior becomes increasingly important.
Generative Hybrid Representations for Activity Forecasting with No-Regret Learning
Automatically reasoning about future human behaviors is a difficult problem but has significant practical applications to assistive systems.
Future Near-Collision Prediction from Monocular Video: Feasibility, Dataset, and Challenges
Our results show that our proposed multi-stream CNN is the best model for predicting time to near-collision.
Robotics
Learnable Embedding Space for Efficient Neural Architecture Compression
We also demonstrate that the learned embedding space can be transferred to new settings for architecture search, such as a larger teacher network or a teacher network in a different architecture family, without any training.
Adversarial Domain Randomization
Domain Randomization (DR) is known to require a significant amount of training data for good performance.
Domain Randomization for Scene-Specific Car Detection and Pose Estimation
We address the issue of domain gap when making use of synthetic data to train a scene-specific object detector and pose estimator.
Directed-Info GAIL: Learning Hierarchical Policies from Unsegmented Demonstrations using Directed Information
The use of imitation learning to learn a single policy for a complex task that has multiple modes or hierarchical structure can be challenging.
R2P2: A ReparameteRized Pushforward Policy for Diverse, Precise Generative Path Forecasting
We propose a method to forecast a vehicle's ego-motion as a distribution over spatiotemporal paths, conditioned on features (e. g., from LIDAR and images) embedded in an overhead map.
Error Correction Maximization for Deep Image Hashing
We propose to use the concept of the Hamming bound to derive the optimal criteria for learning hash codes with a deep network.
Human-Interactive Subgoal Supervision for Efficient Inverse Reinforcement Learning
The learning process is interactive, with a human expert first providing input in the form of full demonstrations along with some subgoal states.
Learning Neural Parsers with Deterministic Differentiable Imitation Learning
We introduce a novel deterministic policy gradient update, DRAG (i. e., DeteRministically AGgrevate) in the form of a deterministic actor-critic variant of AggreVaTeD, to train our neural parser.
ADA: A Game-Theoretic Perspective on Data Augmentation for Object Detection
We aim to find an optimal adversarial perturbations of the ground truth data (i. e., the worst case perturbations) that forces the object bounding box predictor to learn from the hardest distribution of perturbed examples for better test-time performance.
Efficient K-Shot Learning with Regularized Deep Networks
To efficiently search for optimal groupings conditioned on the input data, we propose a reinforcement learning search strategy using recurrent networks to learn the optimal group assignments for each network layer.
Predictive-State Decoders: Encoding the Future into Recurrent Networks
We seek to combine the advantages of RNNs and PSRs by augmenting existing state-of-the-art recurrent neural networks with Predictive-State Decoders (PSDs), which add supervision to the network's internal state representation to target predicting future observations.
Inverse Reinforcement Learning with Conditional Choice Probabilities
We make an important connection to existing results in econometrics to describe an alternative formulation of inverse reinforcement learning (IRL).
N2N Learning: Network to Network Compression via Policy Gradient Reinforcement Learning
Our approach takes a larger `teacher' network as input and outputs a compressed `student' network derived from the `teacher' network.
Privacy-Preserving Visual Learning Using Doubly Permuted Homomorphic Encryption
We propose a privacy-preserving framework for learning visual classifiers by leveraging distributed private image data.
First-Person Activity Forecasting with Online Inverse Reinforcement Learning
We address the problem of incrementally modeling and forecasting long-term goals of a first-person camera wearer: what the user will do, where they will go, and what goal they seek.
Deep Supervised Hashing with Triplet Labels
The current state-of-the-art deep hashing method DPSH~\cite{li2015feature}, which is based on pairwise labels, performs image feature learning and hash code learning simultaneously by maximizing the likelihood of pairwise similarities.
Gesture-based Bootstrapping for Egocentric Hand Segmentation
Concretely, our approach uses two convolutional neural networks: (1) a gesture network that uses pre-defined motion information to detect the hand region; and (2) an appearance network that learns a person specific model of the hand region based on the output of the gesture network.
Contextual Visual Similarity
Given a query image, a second positive image and a third negative image, dissimilar to the first two images, we define a contextualized similarity search criteria.
In Teacher We Trust: Learning Compressed Models for Pedestrian Detection
It has been observed that many of the parameters of a large network are redundant, allowing for the possibility of learning a smaller network that mimics the outputs of the large network through a process called Knowledge Distillation.
Ego-Surfing: Person Localization in First-Person Videos Using Ego-Motion Signatures
We envision a future time when wearable cameras are worn by the masses and recording first-person point-of-view videos of everyday life.
Recognizing Micro-Actions and Reactions From Paired Egocentric Videos
We aim to understand the dynamics of social interactions between two people by recognizing their actions and reactions using a head-mounted camera.
Going Deeper into First-Person Activity Recognition
Our appearance stream encodes prior knowledge of the egocentric paradigm by explicitly training the network to segment hands and localize objects.
Learning Action Maps of Large Environments via First-Person Vision
When people observe and interact with physical spaces, they are able to associate functionality to regions in the environment.
Forecasting Interactive Dynamics of Pedestrians with Fictitious Play
We develop predictive models of pedestrian dynamics by encoding the coupled nature of multi-pedestrian interaction using game theory, and deep learning-based visual analysis to estimate person-specific behavior parameters.
Learning Scene-Specific Pedestrian Detectors Without Real Data
Our results also yield a surprising result, that our method using purely synthetic data is able to outperform models trained on real scene-specific data when data is limited.
How Do We Use Our Hands? Discovering a Diverse Set of Common Grasps
Furthermore, we develop a hierarchical extension to the DPP clustering algorithm and show that it can be used to discover appearance-based grasp taxonomies.
Ego-Surfing First-Person Videos
We incorporate this feature into our proposed approach that computes the motion correlation over supervoxel hierarchies to localize target instances in observer videos.
Pixel-Level Hand Detection in Ego-centric Videos
Our analysis highlights the effectiveness of sparse features and the importance of modeling global illumination.
