137 citations · 224 across the 13 of their papers we have counts for
14 papers
SEGIC: Unleashing the Emergent Correspondence for In-Context Segmentation
Lingchen Meng, Shiyi Lan, Hengduo Li +3
In-context segmentation aims at segmenting novel images using a few labeled example images, termed as "in-context examples", exploring content similarities between examples and the…
BMB: Balanced Memory Bank for Imbalanced Semi-supervised Learning
Wujian Peng, Zejia Weng, Hengduo Li +1
Exploring a substantial amount of unlabeled data, semi-supervised learning (SSL) boosts the recognition performance when only a limited number of labels are provided. However, trad…
Semi-Supervised Single-View 3D Reconstruction via Prototype Shape Priors
Zhen Xing, Hengduo Li, Zuxuan Wu +1
The performance of existing single-view 3D reconstruction methods heavily relies on large-scale 3D annotations. However, such annotations are tedious and expensive to collect. Semi…
AdaViT: Adaptive Vision Transformers for Efficient Image Recognition
Lingchen Meng, Hengduo Li, Bor-Chun Chen +4
Built on top of self-attention mechanisms, vision transformers have demonstrated remarkable performance on a variety of vision tasks recently. While achieving excellent performance…
Efficient Video Transformers with Spatial-Temporal Token Selection
Junke Wang, Xitong Yang, Hengduo Li +3
Video transformers have achieved impressive results on major video recognition benchmarks, which however suffer from high computational cost. In this paper, we present STTS, a toke…
Rethinking Pseudo Labels for Semi-Supervised Object Detection
Hengduo Li, Zuxuan Wu, Abhinav Shrivastava +1
Recent advances in semi-supervised object detection (SSOD) are largely driven by consistency-based pseudo-labeling methods for image classification tasks, producing pseudo labels a…
HCMS: Hierarchical and Conditional Modality Selection for Efficient Video Recognition
Zejia Weng, Zuxuan Wu, Hengduo Li +2
Videos are multimodal in nature. Conventional video recognition pipelines typically fuse multimodal features for improved performance. However, this is not only computationally exp…
2D or not 2D? Adaptive 3D Convolution Selection for Efficient Video Recognition
Hengduo Li, Zuxuan Wu, Abhinav Shrivastava +1
3D convolutional networks are prevalent for video recognition. While achieving excellent recognition performance on standard benchmarks, they operate on a sequence of frames with 3…
Improving the Tightness of Convex Relaxation Bounds for Training Certifiably Robust Classifiers
Chen Zhu, Renkun Ni, Ping-yeh Chiang +3
Convex relaxations are effective for training and certifying neural networks against norm-bounded adversarial attacks, but they leave a large gap between certifiable and empirical…
Learning from Noisy Anchors for One-stage Object Detection
Hengduo Li, Zuxuan Wu, Chen Zhu +3
State-of-the-art object detectors rely on regressing and classifying an extensive list of possible anchors, which are divided into positive and negative samples based on their inte…
Transferable Clean-Label Poisoning Attacks on Deep Neural Nets
Chen Zhu, W. Ronny Huang, Ali Shafahi +4
Clean-label poisoning attacks inject innocuous looking (and "correctly" labeled) poison images into training data, causing a model to misclassify a targeted image after being train…
An Analysis of Pre-Training on Object Detection
Hengduo Li, Bharat Singh, Mahyar Najibi +2
We provide a detailed analysis of convolutional neural networks which are pre-trained on the task of object detection. To this end, we train detectors on large datasets like OpenIm…
R-FCN-3000 at 30fps: Decoupling Detection and Classification
Bharat Singh, Hengduo Li, Abhishek Sharma +1
We present R-FCN-3000, a large-scale real-time object detector in which objectness detection and classification are decoupled. To obtain the detection score for an RoI, we multiply…
Multi-Glimpse LSTM with Color-Depth Feature Fusion for Human Detection
Hengduo Li, Jun Liu, Guyue Zhang +2
With the development of depth cameras such as Kinect and Intel Realsense, RGB-D based human detection receives continuous research attention due to its usage in a variety of applic…