Search Results for author:
Found 51 papers, 23 papers with code
Tensor-Based Human Body Modeling
In this paper, we present a novel approach to model 3D human body with variations on both human shape and pose, by exploring a tensor decomposition technique.
Reinforced Temporal Attention and Split-Rate Transfer for Depth-Based Person Re-Identification
We address the problem of person re-identification from commodity depth sensors.
Incremental Classifier Learning with Generative Adversarial Networks
To address these problems, we propose (a) a new loss function to combine the cross-entropy loss and distillation loss, (b) a simple way to estimate and remove the unbalance between the old and new classes , and (c) using Generative Adversarial Networks (GANs) to generate historical data and select representative exemplars during generation.
Rethinking Classification and Localization for Object Detection
Two head structures (i. e. fully connected head and convolution head) have been widely used in R-CNN based detectors for classification and localization tasks.
Large Scale Incremental Learning
We believe this is because of the combination of two factors: (a) the data imbalance between the old and new classes, and (b) the increasing number of visually similar classes.
Cross-Domain Complementary Learning Using Pose for Multi-Person Part Segmentation
On the other hand, if part labels are also available in the real-images during training, our method outperforms the supervised state-of-the-art methods by a large margin.
Ranked #1 on
Human Part Segmentation
on PASCAL-Part
(using extra training data)
Unsupervised Domain Adaptation for Object Detection via Cross-Domain Semi-Supervised Learning
In this work, we show that such adversarial-based methods can only reduce the domain style gap, but cannot address the domain content distribution gap that is shown to be important for object detectors.
Dynamic Convolution: Attention over Convolution Kernels
Light-weight convolutional neural networks (CNNs) suffer performance degradation as their low computational budgets constrain both the depth (number of convolution layers) and the width (number of channels) of CNNs, resulting in limited representation capability.
Ranked #905 on
Image Classification
on ImageNet
DA-NAS: Data Adapted Pruning for Efficient Neural Architecture Search
One common way is searching on a smaller proxy dataset (e. g., CIFAR-10) and then transferring to the target task (e. g., ImageNet).
MicroNet: Towards Image Recognition with Extremely Low FLOPs
In this paper, we present MicroNet, which is an efficient convolutional neural network using extremely low computational cost (e. g. 6 MFLOPs on ImageNet classification).
Are Fewer Labels Possible for Few-shot Learning?
We conduct experiments on 10 different few-shot target datasets, and our average few-shot performance outperforms both vanilla inductive unsupervised transfer and supervised transfer by a large margin.
Dynamic DETR: End-to-End Object Detection With Dynamic Attention
To mitigate the second limitation of learning difficulty, we introduce a dynamic decoder by replacing the cross-attention module with a ROI-based dynamic attention in the Transformer decoder.
Weak NAS Predictor Is All You Need
Rather than expecting a single strong predictor to model the whole space, we seek a progressive line of weak predictors that can connect a path to the best architecture, thus greatly simplifying the learning task of each predictor.
Stronger NAS with Weaker Predictors
We propose a paradigm shift from fitting the whole architecture space using one strong predictor, to progressively fitting a search path towards the high-performance sub-space through a set of weaker predictors.
Revisiting Dynamic Convolution via Matrix Decomposition
It has two limitations: (a) it increases the number of convolutional weights by K-times, and (b) the joint optimization of dynamic attention and static convolution kernels is challenging.
Dynamic Transfer for Multi-Source Domain Adaptation
However, such a static model is difficult to handle conflicts across multiple domains, and suffers from a performance degradation in both source domains and target domain.
Dynamic Head: Unifying Object Detection Heads with Attentions
In this paper, we present a novel dynamic head framework to unify object detection heads with attentions.
Ranked #2 on
Object Detection
on COCO 2017 val
(AP75 metric)
Improve Unsupervised Pretraining for Few-label Transfer
Unsupervised pretraining has achieved great success and many recent works have shown unsupervised pretraining can achieve comparable or even slightly better transfer performance than supervised pretraining on downstream target datasets.
Is Object Detection Necessary for Human-Object Interaction Recognition?
This paper revisits human-object interaction (HOI) recognition at image level without using supervisions of object location and human pose.
Mobile-Former: Bridging MobileNet and Transformer
This structure leverages the advantages of MobileNet at local processing and transformer at global interaction.
MicroNet: Improving Image Recognition with Extremely Low FLOPs
This paper aims at addressing the problem of substantial performance degradation at extremely low computational cost (e. g. 5M FLOPs on ImageNet classification).
Unsupervised Learning of Full-Waveform Inversion: Connecting CNN and Partial Differential Equation in a Loop
In particular, we use finite difference to approximate the forward modeling of PDE as a differentiable operator (from velocity map to seismic data) and model its inversion by CNN (from seismic data to velocity map).
Unsupervised Finetuning
This problem is more challenging than the supervised counterpart, as the low data density in the small-scale target data is not friendly for unsupervised learning, leading to the damage of the pretrained representation and poor representation in the target domain.
OpenFWI: Large-Scale Multi-Structural Benchmark Datasets for Seismic Full Waveform Inversion
The recent success of data-driven FWI methods results in a rapidly increasing demand for open datasets to serve the geophysics community.
BEVT: BERT Pretraining of Video Transformers
This design is motivated by two observations: 1) transformers learned on image datasets provide decent spatial priors that can ease the learning of video transformers, which are often times computationally-intensive if trained from scratch; 2) discriminative clues, i. e., spatial and temporal information, needed to make correct predictions vary among different videos due to large intra-class and inter-class variations.
Ranked #8 on
Action Recognition
on Diving-48
Improving Vision Transformers for Incremental Learning
This paper proposes a working recipe of using Vision Transformer (ViT) in class incremental learning.
The Overlooked Classifier in Human-Object Interaction Recognition
Human-Object Interaction (HOI) recognition is challenging due to two factors: (1) significant imbalance across classes and (2) requiring multiple labels per image.
Ranked #1 on
Human-Object Interaction Detection
on HICO
Extremely Weak Supervision Inversion of Multi-physical Properties
We explore a multi-physics inversion problem from two distinct measurements~(seismic and EM data) to three geophysical properties~(velocity, conductivity, and CO$_2$ saturation).
The Overlooked Classifier in Human-Object Interaction Recognition
Human-Object Interaction (HOI) recognition is challenging due to two factors: (1) significant imbalance across classes and (2) requiring multiple labels per image.
Residual Mixture of Experts
In this paper, we propose Residual Mixture of Experts (RMoE), an efficient training pipeline for MoE vision transformers on downstream tasks, such as segmentation and detection.
An Intriguing Property of Geophysics Inversion
In particular, when dealing with the inversion from seismic data to subsurface velocity governed by a wave equation, the integral results of velocity with Gaussian kernels are linearly correlated to the integral of seismic data with sine kernels.
Reduce Information Loss in Transformers for Pluralistic Image Inpainting
The indices of quantized pixels are used as tokens for the inputs and prediction targets of transformer.
Ranked #6 on
Seeing Beyond the Visible
on KITTI360-EX
Detection Hub: Unifying Object Detection Datasets via Query Adaptation on Language Embedding
Detection Hub further achieves SoTA performance on UODB benchmark with wide variety of datasets.
Should All Proposals be Treated Equally in Object Detection?
We formulate this as a learning problem where the goal is to assign operators to proposals, in the detection head, so that the total computational cost is constrained and the precision is maximized.
Design What You Desire: Icon Generation from Orthogonal Application and Theme Labels
Generative adversarial networks (GANs) have been trained to be professional artists able to create stunning artworks such as face generation and image style transfer.
Video Mobile-Former: Video Recognition with Efficient Global Spatial-temporal Modeling
To avoid significant computational cost incurred by computing self-attention between the large number of local patches in videos, we propose to use very few global tokens (e. g., 6) for a whole video in Transformers to exchange information with 3D-CNNs with a cross-attention mechanism.
Self-Supervised Learning based on Heat Equation
When transferring to object detection with frozen backbone, QB-Heat outperforms MoCo-v2 and supervised pre-training on ImageNet by 7. 9 and 4. 5 AP respectively.
Masked Video Distillation: Rethinking Masked Feature Modeling for Self-supervised Video Representation Learning
For the choice of teacher models, we observe that students taught by video teachers perform better on temporally-heavy video tasks, while image teachers transfer stronger spatial representations for spatially-heavy video tasks.
Ranked #1 on
Self-Supervised Action Recognition
on HMDB51
Find Beauty in the Rare: Contrastive Composition Feature Clustering for Nontrivial Cropping Box Regression
Observing that similar composition patterns tend to be shared by the cropping boundaries annotated nearly, we argue to find the beauty of composition from the rare samples by clustering the samples with similar cropping boundary annotations, ie, similar composition patterns.
Layer Grafted Pre-training: Bridging Contrastive Learning And Masked Image Modeling For Label-Efficient Representations
This motivates us to shift the paradigm from combining loss at the end, to choosing the proper learning method per network layer.
Simplifying Full Waveform Inversion via Domain-Independent Self-Supervised Learning
Geophysics has witnessed success in applying deep learning to one of its core problems: full waveform inversion (FWI) to predict subsurface velocity maps from seismic data.
Image as First-Order Norm+Linear Autoregression: Unveiling Mathematical Invariance
This paper introduces a novel mathematical property applicable to diverse images, referred to as FINOLA (First-Order Norm+Linear Autoregressive).
PaintSeg: Training-free Segmentation via Painting
The paper introduces PaintSeg, a new unsupervised method for segmenting objects without any training.
Designing a Better Asymmetric VQGAN for StableDiffusion
The training cost of our asymmetric VQGAN is cheap, and we only need to retrain a new asymmetric decoder while keeping the vanilla VQGAN encoder and StableDiffusion unchanged.
$\mathbf{\mathbb{E}^{FWI}}$: Multi-parameter Benchmark Datasets for Elastic Full Waveform Inversion of Geophysical Properties
We conduct comprehensive numerical experiments to explore the relationship between P-wave and S-wave velocities in seismic data.
Does Full Waveform Inversion Benefit from Big Data?
This paper investigates the impact of big data on deep learning models for full waveform inversion (FWI).
Improving Adversarial Robustness of Masked Autoencoders via Test-time Frequency-domain Prompting
Based on our analysis, we provide a simple yet effective way to boost the adversarial robustness of MAE.
Completing Visual Objects via Bridging Generation and Segmentation
This paper presents a novel approach to object completion, with the primary goal of reconstructing a complete object from its partially visible components.
On the Hidden Waves of Image
We term this phenomenon hidden waves, as it reveals that, although the speeds of the set of wave equations and autoregressive coefficient matrices are latent, they are both learnable and shared across images.
Efficient Modulation for Vision Networks
We demonstrate that the modulation mechanism is particularly well suited for efficient networks and further tailor the modulation design by proposing the efficient modulation (EfficientMod) block, which is considered the essential building block for our networks.
