• Publications
  • Influence
SSH: Single Stage Headless Face Detector
We introduce the Single Stage Headless (SSH) face detector. Unlike two stage proposal-classification detectors, SSH detects faces in a single stage directly from the early convolutional layers in aExpand
  • 219
  • 46
  • PDF
Poison Frogs! Targeted Clean-Label Poisoning Attacks on Neural Networks
TLDR
Data poisoning is an attack on machine learning models wherein the attacker adds examples to the training set to manipulate the behavior of the model at test time. Expand
  • 255
  • 26
  • PDF
SNIPER: Efficient Multi-Scale Training
TLDR
We present SNIPER, an algorithm for performing efficient multi-scale training in instance level visual recognition tasks, which adaptively samples chips from multiple scales in an image pyramid. Expand
  • 197
  • 26
  • PDF
Adversarial Training for Free!
TLDR
Adversarial training is time-consuming—in addition to the gradient computation needed to update the network parameters, each stochastic gradient descent (SGD) iteration requires multiple gradient computations to produce adversarial images. Expand
  • 249
  • 25
  • PDF
Universal Adversarial Training
TLDR
We propose universal adversarial training, which models the problem of robust classifier generation as a two-player min-max game, and produces robust models with only 2X the cost of natural training. Expand
  • 47
  • 11
  • PDF
G-CNN: An Iterative Grid Based Object Detector
TLDR
We introduce G-CNN, an object detection technique based on CNNs which works without proposal algorithms and achieves comparable results to state-of-the-art. Expand
  • 94
  • 6
  • PDF
Soft Sampling for Robust Object Detection
TLDR
We study the robustness of object detection under the presence of missing annotations. Expand
  • 27
  • 5
  • PDF
AutoFocus: Efficient Multi-Scale Inference
TLDR
This paper describes AutoFocus, an efficient multi-scale inference algorithm for deep-learning based object detectors. Expand
  • 36
  • 2
  • PDF
FA-RPN: Floating Region Proposals for Face Detection
TLDR
We propose a novel approach for generating region proposals for performing face detection. Expand
  • 17
  • 1
  • PDF
Generate, Segment and Replace: Towards Generic Manipulation Segmentation
TLDR
We propose an end-to-end manipulation segmentation network based on semantic segmentation for detecting image manipulation. Expand
  • 8
  • 1
  • PDF
...
1
2
3
...