There is extensive literature on perceiving road structures by fusing va...
Multi-task learning has emerged as a powerful paradigm to solve a range ...
Large-scale cross-modal pre-training paradigms have recently shown ubiqu...
Inspired by the success of visual-language methods (VLMs) in zero-shot
c...
Open-world object detection, as a more general and challenging goal, aim...
Aiming towards a holistic understanding of multiple downstream tasks
sim...
To bridge the gap between supervised semantic segmentation and real-worl...
Self-supervised learning (SSL), especially contrastive methods, has rais...
Existing text-guided image manipulation methods aim to modify the appear...
We present Laneformer, a conceptually simple yet powerful transformer-ba...
Contemporary deep-learning object detection methods for autonomous drivi...
This paper presents a large-scale Chinese cross-modal dataset for
benchm...
Deploying convolutional neural networks (CNNs) on mobile devices is diff...
Adder neural networks (AdderNets) have shown impressive performance on i...
Unsupervised large-scale vision-language pre-training has shown promisin...
Vision transformers (ViTs) have pushed the state-of-the-art for various
...
This paper introduces versatile filters to construct efficient convoluti...
We present a flexible and high-performance framework, named Pyramid R-CN...
We present Voxel Transformer (VoTr), a novel and effective voxel-based
T...
Recent studies on deep convolutional neural networks present a simple
pa...
Vision transformers have been successfully applied to image recognition ...
Shift neural networks reduce computation complexity by removing expensiv...
Aiming at facilitating a real-world, ever-evolving and scalable autonomo...
Current perception models in autonomous driving have become notorious fo...
Adder neural network (AdderNet) is a new kind of deep model that replace...
Knowledge distillation is a widely used paradigm for inheriting informat...
Deep learning based methods, especially convolutional neural networks (C...
Binary neural networks (BNNs) represent original full-precision weights ...
Transformer is a type of self-attention-based neural networks originally...
Convolutional neural networks (CNN) have been widely used for boosting t...
Modern single image super-resolution (SISR) system based on convolutiona...
Transformer is a type of deep neural network mainly based on self-attent...
As the computing power of modern hardware is increasing strongly, pre-tr...
This paper proposes a reliable neural network pruning algorithm by setti...
This paper formalizes the binarization operations over neural networks f...
Adder Neural Networks (ANNs) which only contain additions bring us a new...
This paper studies the single image super-resolution problem using adder...
Quantized neural networks with low-bit weights and activations are attra...
Video super-resolution, which aims at producing a high-resolution video ...
This paper focuses on channel pruning for semantic segmentation networks...
Video style transfer techniques inspire many exciting applications on mo...
Neural Architecture Search (NAS) refers to automatically design the
arch...
Neural architecture search (NAS) aims to automatically design deep neura...
Despite Generative Adversarial Networks (GANs) have been widely used in
...
Deep neural networks often consist of a great number of trainable parame...
Quantization neural networks (QNNs) are very attractive to the industry
...
Compared with cheap addition operation, multiplication operation is of m...
Deploying convolutional neural networks (CNNs) on embedded devices is
di...
The task of single image super-resolution (SISR) aims at reconstructing ...
Neural Architecture Search (NAS) is attractive for automatically produci...