Search Results for author:
Found 78 papers, 45 papers with code
Fast and Balanced: Efficient Label Tree Learning for Large Scale Object Recognition
We present a novel approach to efficiently learn a label tree for large scale classification with many classes.
Fine-Grained Crowdsourcing for Fine-Grained Recognition
In this work, we include humans in the loop to help computers select discriminative features.
ImageNet Large Scale Visual Recognition Challenge
The ImageNet Large Scale Visual Recognition Challenge is a benchmark in object category classification and detection on hundreds of object categories and millions of images.
Probabilistic Label Relation Graphs with Ising Models
In this paper, we extend the HEX model to allow for soft or probabilistic relations between labels, which is useful when there is uncertainty about the relationship between two labels (e. g., an antelope is "sort of" furry, but not to the same degree as a grizzly bear).
Learning Semantic Relationships for Better Action Retrieval in Images
Human actions capture a wide variety of interactions between people and objects.
Mining Semantic Affordances of Visual Object Categories
In this paper we introduce the new problem of mining the knowledge of semantic affordance: given an object, determining whether an action can be performed on it.
HICO: A Benchmark for Recognizing Human-Object Interactions in Images
We introduce a new benchmark "Humans Interacting with Common Objects" (HICO) for recognizing human-object interactions (HOI).
Stacked Hourglass Networks for Human Pose Estimation
This work introduces a novel convolutional network architecture for the task of human pose estimation.
Ranked #1 on
Pose Estimation
on FLIC Wrists
Single-Image Depth Perception in the Wild
This paper studies single-image depth perception in the wild, i. e., recovering depth from a single image taken in unconstrained settings.
Associative Embedding: End-to-End Learning for Joint Detection and Grouping
We introduce associative embedding, a novel method for supervising convolutional neural networks for the task of detection and grouping.
Ranked #5 on
Keypoint Detection
on OCHuman
Dynamic Deep Neural Networks: Optimizing Accuracy-Efficiency Trade-offs by Selective Execution
We introduce Dynamic Deep Neural Networks (D2NN), a new type of feed-forward deep neural network that allows selective execution.
Learning to Detect Human-Object Interactions
We study the problem of detecting human-object interactions (HOI) in static images, defined as predicting a human and an object bounding box with an interaction class label that connects them.
General Classification
Human-Object Interaction Detection
+1
Using Deep Learning and Google Street View to Estimate the Demographic Makeup of the US
The United States spends more than $1B each year on initiatives such as the American Community Survey (ACS), a labor-intensive door-to-door study that measures statistics relating to race, gender, education, occupation, unemployment, and other demographic factors.
Surface Normals in the Wild
We study the problem of single-image depth estimation for images in the wild.
Forecasting Human Dynamics from Static Images
This paper presents the first study on forecasting human dynamics from static images.
Temporal Action Localization by Structured Maximal Sums
We pose action localization as a structured prediction over arbitrary-length temporal windows, where each window is scored as the sum of frame-wise classification scores.
Fine-Grained Car Detection for Visual Census Estimation
In this work, we leverage the ubiquity of Google Street View images and develop a computer vision pipeline to predict income, per capita carbon emission, crime rates and other city attributes from a single source of publicly available visual data.
Scalable Annotation of Fine-Grained Categories Without Experts
We present a crowdsourcing workflow to collect image annotations for visually similar synthetic categories without requiring experts.
Premise Selection for Theorem Proving by Deep Graph Embedding
We propose a deep learning-based approach to the problem of premise selection: selecting mathematical statements relevant for proving a given conjecture.
Ranked #1 on
Automated Theorem Proving
on HolStep (Unconditional)
Shape from Shading through Shape Evolution
The evolution generates better shapes guided by the network training, while the training improves by using the evolved shapes.
Rethinking the Faster R-CNN Architecture for Temporal Action Localization
We propose TAL-Net, an improved approach to temporal action localization in video that is inspired by the Faster R-CNN object detection framework.
Ranked #27 on
Temporal Action Localization
on THUMOS’14
Decorrelated Batch Normalization
Batch Normalization (BN) is capable of accelerating the training of deep models by centering and scaling activations within mini-batches.
Think Visually: Question Answering through Virtual Imagery
In this paper, we study the problem of geometric reasoning in the context of question-answering.
Learning Single-Image Depth from Videos using Quality Assessment Networks
Depth estimation from a single image in the wild remains a challenging problem.
CornerNet: Detecting Objects as Paired Keypoints
We propose CornerNet, a new approach to object detection where we detect an object bounding box as a pair of keypoints, the top-left corner and the bottom-right corner, using a single convolution neural network.
Ranked #172 on
Object Detection
on COCO test-dev
Rethinking Numerical Representations for Deep Neural Networks
With ever-increasing computational demand for deep learning, it is critical to investigate the implications of the numeric representation and precision of DNN model weights and activations on computational efficiency.
Speaker Naming in Movies
We propose a new model for speaker naming in movies that leverages visual, textual, and acoustic modalities in an unified optimization framework.
MeshAdv: Adversarial Meshes for Visual Recognition
Highly expressive models such as deep neural networks (DNNs) have been widely applied to various applications.
DeepV2D: Video to Depth with Differentiable Structure from Motion
We propose DeepV2D, an end-to-end deep learning architecture for predicting depth from video.
D3D: Distilled 3D Networks for Video Action Recognition
State-of-the-art methods for video action recognition commonly use an ensemble of two networks: the spatial stream, which takes RGB frames as input, and the temporal stream, which takes optical flow as input.
Ranked #11 on
Action Recognition
on AVA v2.1
CornerNet-Lite: Efficient Keypoint Based Object Detection
Together these two variants address the two critical use cases in efficient object detection: improving efficiency without sacrificing accuracy, and improving accuracy at real-time efficiency.
Ranked #137 on
Object Detection
on COCO minival
Learning to Prove Theorems via Interacting with Proof Assistants
Proof assistants offer a formalism that resembles human mathematical reasoning, representing theorems in higher-order logic and proofs as high-level tactics.
Ranked #1 on
Automated Theorem Proving
on CoqGym
Identifying Visible Actions in Lifestyle Vlogs
We consider the task of identifying human actions visible in online videos.
Learning to Generate Synthetic 3D Training Data through Hybrid Gradient
We parametrize the design decisions as a real vector, and combine the approximate gradient and the analytical gradient to obtain the hybrid gradient of the network performance with respect to this vector.
To Learn or Not to Learn: Analyzing the Role of Learning for Navigation in Virtual Environments
In this paper we compare learning-based methods and classical methods for navigation in virtual environments.
SpatialSense: An Adversarially Crowdsourced Benchmark for Spatial Relation Recognition
Understanding the spatial relations between objects in images is a surprisingly challenging task.
Feature Partitioning for Efficient Multi-Task Architectures
Multi-task learning holds the promise of less data, parameters, and time than training of separate models.
Learning to Sit: Synthesizing Human-Chair Interactions via Hierarchical Control
We experimentally demonstrate the strength of our approach over different non-hierarchical and hierarchical baselines.
Generative Modeling for Small-Data Object Detection
This paper explores object detection in the small data regime, where only a limited number of annotated bounding boxes are available due to data rarity and annotation expense.
Representing Movie Characters in Dialogues
We introduce a new embedding model to represent movie characters and their interactions in a dialogue by encoding in the same representation the language used by these characters as well as information about the other participants in the dialogue.
Compositional Temporal Visual Grounding of Natural Language Event Descriptions
Temporal grounding entails establishing a correspondence between natural language event descriptions and their visual depictions.
Towards Fairer Datasets: Filtering and Balancing the Distribution of the People Subtree in the ImageNet Hierarchy
Computer vision technology is being used by many but remains representative of only a few.
Learning to Prove Theorems by Learning to Generate Theorems
We consider the task of automated theorem proving, a key AI task.
Ranked #2 on
Automated Theorem Proving
on Metamath set.mm
RAFT: Recurrent All-Pairs Field Transforms for Optical Flow
RAFT extracts per-pixel features, builds multi-scale 4D correlation volumes for all pairs of pixels, and iteratively updates a flow field through a recurrent unit that performs lookups on the correlation volumes.
Ranked #6 on
Optical Flow Estimation
on Spring
How Useful is Self-Supervised Pretraining for Visual Tasks?
We investigate what factors may play a role in the utility of these pretraining methods for practitioners.
LifeQA: A Real-life Dataset for Video Question Answering
We introduce LifeQA, a benchmark dataset for video question answering that focuses on day-to-day real-life situations.
PackIt: A Virtual Environment for Geometric Planning
The ability to jointly understand the geometry of objects and plan actions for manipulating them is crucial for intelligent agents.
Ranked #1 on
Robot Task Planning
on PackIt
OASIS: A Large-Scale Dataset for Single Image 3D in the Wild
Single-view 3D is the task of recovering 3D properties such as depth and surface normals from a single image.
A Unified Framework of Surrogate Loss by Refactoring and Interpolation
We introduce UniLoss, a unified framework to generate surrogate losses for training deep networks with gradient descent, reducing the amount of manual design of task-specific surrogate losses.
Learning Video Representations from Textual Web Supervision
Based on this observation, we propose to use text as a method for learning video representations.
Strongly Incremental Constituency Parsing with Graph Neural Networks
Based on our transition system, we develop a strongly incremental parser.
Ranked #1 on
Constituency Parsing
on CTB5
Rearrangement: A Challenge for Embodied AI
In the rearrangement task, the goal is to bring a given physical environment into a specified state.
RAFT-3D: Scene Flow using Rigid-Motion Embeddings
We address the problem of scene flow: given a pair of stereo or RGB-D video frames, estimate pixelwise 3D motion.
Ranked #2 on
Scene Flow Estimation
on Spring
Rel3D: A Minimally Contrastive Benchmark for Grounding Spatial Relations in 3D
The 3D scenes in our dataset come in minimally contrastive pairs: two scenes in a pair are almost identical, but a spatial relation holds in one and fails in the other.
Ranked #1 on
Spatial Relation Recognition
on Rel3D
Revisiting Point Cloud Classification with a Simple and Effective Baseline
It also outperforms state-of-the-art methods on ScanObjectNN, a real-world point cloud benchmark, and demonstrates better cross-dataset generalization.
Ranked #11 on
3D Point Cloud Classification
on ModelNet40-C
HYPE-C: Evaluating Image Completion Models Through Standardized Crowdsourcing
Recent work has proposed the use of human evaluation for image synthesis models, allowing for a reliable method to evaluate the visual quality of generated images.
Tangent Space Backpropagation for 3D Transformation Groups
We address the problem of performing backpropagation for computation graphs involving 3D transformation groups SO(3), SE(3), and Sim(3).
Revisiting Point Cloud Shape Classification with a Simple and Effective Baseline
It also outperforms state-of-the-art methods on ScanObjectNN, a real-world point cloud benchmark, and demonstrates better cross-dataset generalization.
Ranked #18 on
Point Cloud Classification
on PointCloud-C
Dynamically Grown Generative Adversarial Networks
Recent work introduced progressive network growing as a promising way to ease the training for large GANs, but the model design and architecture-growing strategy still remain under-explored and needs manual design for different image data.
DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Cameras
We introduce DROID-SLAM, a new deep learning based SLAM system.
RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching
We introduce RAFT-Stereo, a new deep architecture for rectified stereo based on the optical flow network RAFT.
Ranked #1 on
Stereo Disparity Estimation
on Middlebury 2014
Non-deep Networks
This begs the question -- is it possible to build high-performing "non-deep" neural networks?
Learning Symbolic Rules for Reasoning in Quasi-Natural Language
In this work, we ask how we can build a rule-based system that can reason with natural language input but without the manual construction of rules.
IFOR: Iterative Flow Minimization for Robotic Object Rearrangement
Accurate object rearrangement from vision is a crucial problem for a wide variety of real-world robotics applications in unstructured environments.
Coupled Iterative Refinement for 6D Multi-Object Pose Estimation
We propose a new approach to 6D object pose estimation which consists of an end-to-end differentiable architecture that makes use of geometric knowledge.
Multiview Stereo with Cascaded Epipolar RAFT
CER-MVS is significantly different from prior work in multiview stereo.
View Synthesis with Sculpted Neural Points
In this work, we propose a new approach that performs view synthesis using point clouds.
Generating Natural Language Proofs with Verifier-Guided Search
In this paper, we present a novel stepwise method, NLProofS (Natural Language Proof Search), which learns to generate relevant steps conditioning on the hypothesis.
Label-Free Synthetic Pretraining of Object Detectors
Our "SOLID" approach consists of two main components: (1) generating synthetic images using a collection of unlabelled 3D models with optimized scene arrangement; (2) pretraining an object detector on "instance detection" task - given a query image depicting an object, detecting all instances of the exact same object in a target image.
Deep Patch Visual Odometry
DPVO disproves this assumption, showing that it is possible to get the best accuracy and efficiency by exploiting the advantages of sparse patch-based matching over dense flow.
Infinite Photorealistic Worlds using Procedural Generation
We introduce Infinigen, a procedural generator of photorealistic 3D scenes of the natural world.
Convolutional Networks with Oriented 1D Kernels
Specifically, we find that one key ingredient to a high-performing 1D ConvNet is oriented 1D kernels: 1D kernels that are oriented not just horizontally or vertically, but also at other angles.
Llemma: An Open Language Model For Mathematics
We present Llemma, a large language model for mathematics.
Ranked #5 on
Automated Theorem Proving
on miniF2F-test
View-Dependent Octree-based Mesh Extraction in Unbounded Scenes for Procedural Synthetic Data
Procedural synthetic data generation has received increasing attention in computer vision.
Tell Me More! Towards Implicit User Intention Understanding of Language Model Driven Agents
Current language model-driven agents often lack mechanisms for effective user participation, which is crucial given the vagueness commonly found in user instructions.
Advancing LLM Reasoning Generalists with Preference Trees
We introduce Eurus, a suite of large language models (LLMs) optimized for reasoning.
