Search Results for author:
Found 57 papers, 25 papers with code
PeLK: Parameter-efficient Large Kernel ConvNets with Peripheral Convolution
Due to these issues, current CNNs compromise to scale up to 51x51 in the form of stripe convolution (i. e., 51x5 + 5x51) and start to saturate as the kernel size continues growing.
Safe Reinforcement Learning with Free-form Natural Language Constraints and Pre-Trained Language Models
Through the use of pre-trained LMs and the elimination of the need for a ground-truth cost, our method enhances safe policy learning under a diverse set of human-derived free-form natural language constraints.
TAPE: Leveraging Agent Topology for Cooperative Multi-Agent Policy Gradient
We propose Topology-based multi-Agent Policy gradiEnt (TAPE) for both stochastic and deterministic MAPG methods.
Efficient-VQGAN: Towards High-Resolution Image Generation with Efficient Vision Transformers
Vector-quantized image modeling has shown great potential in synthesizing high-quality images.
See Your Heart: Psychological states Interpretation through Visual Creations
Dataset analysis illustrates that SpyIn is not only able to support VEIT, but also more challenging compared with other captioning datasets.
PECAN: Leveraging Policy Ensemble for Context-Aware Zero-Shot Human-AI Coordination
We conduct experiments on the Overcooked environment, and evaluate the zero-shot human-AI coordination performance of our method with both behavior-cloned human proxies and real humans.
InsPro: Propagating Instance Query and Proposal for Online Video Instance Segmentation
When using such a pair to predict an object instance on the current frame, not only the generated instance is automatically associated with its precursors on previous frames, but the model gets a good prior for predicting the same object.
Learning Disentangled Label Representations for Multi-label Classification
The specificity of the framework lies in a feature disentangle module, which contains learnable semantic queries and a Semantic Spatial Cross-Attention (SSCA) module.
Distributed Deep Reinforcement Learning: A Survey and A Multi-Player Multi-Agent Learning Toolbox
With the breakthrough of AlphaGo, deep reinforcement learning becomes a recognized technique for solving sequential decision-making problems.
DiffGAR: Model-Agnostic Restoration from Generative Artifacts Using Image-to-Image Diffusion Models
In this work, we first design a group of mechanisms to simulate generative artifacts of popular generators (i. e., GANs, autoregressive models, and diffusion models), given real images.
QueryProp: Object Query Propagation for High-Performance Video Object Detection
The proposed QueryProp contains two propagation strategies: 1) query propagation is performed from sparse key frames to dense non-key frames to reduce the redundant computation on non-key frames; 2) query propagation is performed from previous key frames to the current key frame to improve feature representation by temporal context modeling.
RACA: Relation-Aware Credit Assignment for Ad-Hoc Cooperation in Multi-Agent Deep Reinforcement Learning
Specifically, these methods do not explicitly utilize the relationship between agents and cannot adapt to different sizes of inputs.
PanopticDepth: A Unified Framework for Depth-aware Panoptic Segmentation
To overcome these limitations, we propose a unified framework for the DPS task by applying a dynamic convolution technique to both the PS and depth prediction tasks.
Re-parameterizing Your Optimizers rather than Architectures
For the extreme simplicity of model structure, we focus on a VGG-style plain model and showcase that such a simple model trained with a RepOptimizer, which is referred to as RepOpt-VGG, performs on par with or better than the recent well-designed models.
SOTVerse: A User-defined Task Space of Single Object Tracking
Single object tracking (SOT) research falls into a cycle -- trackers perform well on most benchmarks but quickly fail in challenging scenarios, causing researchers to doubt the insufficient data content and take more effort to construct larger datasets with more challenging situations.
Split Semantic Detection in Sandplay Images
Sandplay image, as an important psychoanalysis carrier, is a visual scene constructed by the client selecting and placing sand objects (e. g., sand, river, human figures, animals, vegetation, buildings, etc.).
Global Instance Tracking: Locating Target More Like Humans
Finally, we design a scientific evaluation procedure using human capabilities as the baseline to judge tracking intelligence.
DecisionHoldem: Safe Depth-Limited Solving With Diverse Opponents for Imperfect-Information Games
It is more common in life than perfect-information game.
AI in Human-computer Gaming: Techniques, Challenges and Opportunities
Through this survey, we 1) compare the main difficulties among different kinds of games and the corresponding techniques utilized for achieving professional human level AIs; 2) summarize the mainstream frameworks and techniques that can be properly relied on for developing AIs for complex human-computer gaming; 3) raise the challenges or drawbacks of current techniques in the successful AIs; and 4) try to point out future trends in human-computer gaming AIs.
Spatial and Semantic Consistency Regularizations for Pedestrian Attribute Recognition
To fully exploit inter-image relations and aggregate human prior in the model learning process, we construct a Spatial and Semantic Consistency (SSC) framework that consists of two complementary regularizations to achieve spatial and semantic consistency for each attribute.
Rethinking of Pedestrian Attribute Recognition: A Reliable Evaluation under Zero-Shot Pedestrian Identity Setting
Second, based on the proposed definition, we expose the limitations of the existing datasets, which violate the academic norm and are inconsistent with the essential requirement of practical industry application.
Learning to Reweight Imaginary Transitions for Model-Based Reinforcement Learning
More specifically, we evaluate the effect of an imaginary transition by calculating the change of the loss computed on the real samples when we use the transition to train the action-value and policy functions.
Model-based Reinforcement Learning
reinforcement-learning
+1
Learning Category- and Instance-Aware Pixel Embedding for Fast Panoptic Segmentation
Panoptic segmentation (PS) is a complex scene understanding task that requires providing high-quality segmentation for both thing objects and stuff regions.
Rethinking of Pedestrian Attribute Recognition: Realistic Datasets with Efficient Method
Despite various methods are proposed to make progress in pedestrian attribute recognition, a crucial problem on existing datasets is often neglected, namely, a large number of identical pedestrian identities in train and test set, which is not consistent with practical application.
Ranked #3 on
Pedestrian Attribute Recognition
on PETA
GlobalTrack: A Simple and Strong Baseline for Long-term Tracking
Specifically, we propose GlobalTrack, a pure global instance search based tracker that makes no assumption on the temporal consistency of the target's positions and scales.
VisDrone-DET2019: The Vision Meets Drone Object Detection in Image Challenge Results
Results of 33 object detection algorithms are presented.
MVP-Net: Multi-view FPN with Position-aware Attention for Deep Universal Lesion Detection
In this paper, we propose to incorporate domain knowledge in clinical practice into the model design of universal lesion detectors.
Ranked #8 on
Medical Object Detection
on DeepLesion
SSAP: Single-Shot Instance Segmentation With Affinity Pyramid
Moreover, incorporating with the learned affinity pyramid, a novel cascaded graph partition module is presented to sequentially generate instances from coarse to fine.
Point Cloud Super Resolution with Adversarial Residual Graph Networks
The key idea of the proposed network is to exploit the local similarity of point cloud and the analogy between LR input and HR output.
Ranked #2 on
Point Cloud Super Resolution
on SHREC15
Graphics Image and Video Processing
SparseMask: Differentiable Connectivity Learning for Dense Image Prediction
In this paper, we aim at automatically searching an efficient network architecture for dense image prediction.
FastFCN: Rethinking Dilated Convolution in the Backbone for Semantic Segmentation
Modern approaches for semantic segmentation usually employ dilated convolutions in the backbone to extract high-resolution feature maps, which brings heavy computation complexity and memory footprint.
Ranked #40 on
Semantic Segmentation
on PASCAL Context
3D Object Detection Using Scale Invariant and Feature Reweighting Networks
On the other hand, our network obtains the useful features and suppresses the features with less information by a SENet module.
EANet: Enhancing Alignment for Cross-Domain Person Re-identification
Person re-identification (ReID) has achieved significant improvement under the single-domain setting.
GOT-10k: A Large High-Diversity Benchmark for Generic Object Tracking in the Wild
(5) Finally, we develop a comprehensive platform for the tracking community that offers full-featured evaluation toolkits, an online evaluation server, and a responsive leaderboard.
Adversarially Occluded Samples for Person Re-Identification
Person re-identification (ReID) is the task of retrieving particular persons across different cameras.
Discriminative Learning of Latent Features for Zero-Shot Recognition
In this work, we retrospect existing methods and demonstrate the necessity to learn discriminative representations for both visual and semantic instances of ZSL.
Fast End-to-End Trainable Guided Filter
To address the problem, we present a novel building block for FCNs, namely guided filtering layer, which is designed for efficiently generating a high-resolution output given the corresponding low-resolution one and a high-resolution guidance map.
Mixed Supervised Object Detection with Robust Objectness Transfer
Different from previous MSD methods that directly transfer the pre-trained object detectors from existing categories to new categories, we propose a more reasonable and robust objectness transfer approach for MSD.
Deep Crisp Boundaries: From Boundaries to Higher-level Tasks
These ConvNet based edge detectors have approached human level performance on standard benchmarks.
Learning Deep Context-aware Features over Body and Latent Parts for Person Re-identification
It is a challenging task due to the large variations in person pose, occlusion, background clutter, etc How to extract powerful features is a fundamental problem in ReID and is still an open problem today.
Ranked #110 on
Person Re-Identification
on Market-1501
MSC: A Dataset for Macro-Management in StarCraft II
We also split MSC into training, validation and test set for the convenience of evaluation and comparison.
A2-RL: Aesthetics Aware Reinforcement Learning for Image Cropping
Image cropping aims at improving the aesthetic quality of images by adjusting their composition.
Deep Crisp Boundaries
Edge detection had made significant progress with the help of deep Convolutional Networks (ConvNet).
Locality-Sensitive Deconvolution Networks With Gated Fusion for RGB-D Indoor Semantic Segmentation
This layer can learn to adjust the contributions of RGB and depth over each pixel for high-performance object recognition.
Ranked #79 on
Semantic Segmentation
on NYU Depth v2
Beyond triplet loss: a deep quadruplet network for person re-identification
In particular, a quadruplet deep network using a margin-based online hard negative mining is proposed based on the quadruplet loss for the person ReID.
GP-GAN: Towards Realistic High-Resolution Image Blending
Concretely, we propose Gaussian-Poisson Equation to formulate the high-resolution image blending problem, which is a joint optimization constrained by the gradient and color information.
Conditional Image Generation
Generative Adversarial Network
+1
A Large-scale Distributed Video Parsing and Evaluation Platform
Visual surveillance systems have become one of the largest data sources of Big Visual Data in real world.
Weakly-supervised Learning of Mid-level Features for Pedestrian Attribute Recognition and Localization
Based on GoogLeNet, firstly, a set of mid-level attribute features are discovered by novelly designed detection layers, where a max-pooling based weakly-supervised object detection technique is used to train these layers with only image-level labels without the need of bounding box annotations of pedestrian attributes.
A Multi-task Deep Network for Person Re-identification
Person re-identification (ReID) focuses on identifying people across different scenes in video surveillance, which is usually formulated as a binary classification task or a ranking task in current person ReID approaches.
ReD-SFA: Relation Discovery Based Slow Feature Analysis for Trajectory Clustering
For spectral embedding/clustering, it is still an open problem on how to construct an relation graph to reflect the intrinsic structures in data.
Deep Aesthetic Quality Assessment with Semantic Information
Human beings often assess the aesthetic quality of an image coupled with the identification of the image's semantic content.
Ranked #5 on
Aesthetics Quality Assessment
on AVA
A Richly Annotated Dataset for Pedestrian Attribute Recognition
RAP has in total 41, 585 pedestrian samples, each of which is annotated with 72 attributes as well as viewpoints, occlusions, body parts information.
Beyond Tree Structure Models: A New Occlusion Aware Graphical Model for Human Pose Estimation
Occlusion is a main challenge for human pose estimation, which is largely ignored in popular tree structure models.
Query Adaptive Similarity Measure for RGB-D Object Recognition
The reasons are in two-fold: (1) existing similarity measures are sensitive to object pose and scale changes, as well as intra-class variations; and (2) effectively fusing RGB and depth cues is still an open problem.
GRSA: Generalized Range Swap Algorithm for the Efficient Optimization of MRFs
al propose the range move algorithms, which are one of the most successful solvers to this problem.
An equalised global graphical model-based approach for multi-camera object tracking
Non-overlapping multi-camera visual object tracking typically consists of two steps: single camera object tracking and inter-camera object tracking.
Deformable Object Matching via Deformation Decomposition based 2D Label MRF
The MRF energy function is derived from the deformation decomposition model.
