• Publications
  • Influence
Recognition using visual phrases
We introduce a dataset suitable for phrasal recognition that uses familiar PASCAL object categories, and demonstrate significant experimental gains resulting from exploiting visual phrases. Expand
Every Picture Tells a Story: Generating Sentences from Images
We describe a system that can compute a score linking an image to a sentence. Expand
30Hz Object Detection with DPM V5
We describe an implementation of the Deformable Parts Model [1] that operates in a user-defined time-frame. Expand
Fast Template Evaluation with Vector Quantization
We describe a method that achieves a substantial end-to-end speedup over the best current methods, without loss of accuracy. Expand
Phrasal Recognition
  • Ali Farhadi, M. Sadeghi
  • Computer Science, Medicine
  • IEEE Transactions on Pattern Analysis and Machine…
  • 1 December 2013
We introduce visual phrases, complex visual composites like "a person riding a horse." Visual phrases often display significantly reduced visual complexity compared to their component objects because the appearance of those objects can change profoundly when they participate in relations. Expand
Poisson Local Color Correction for Image Stitching
A new method for seamless image stitching is presented. The proposed algorithm is a hybrid method which uses optimal seam methods and smoothes the intensity transition between two images by colorExpand
Local Visual Microphones: Improved Sound Extraction from Silent Video
We show that different parts of image vibrate differently and we can improve sound quality by carefully combining local vibrations. Expand
Multi-Representational Learning for Offline Signature Verification using Multi-Loss Snapshot Ensemble of CNNs
We propose a novel multi-loss Snapshot Ensemble framework forOffline Signature Verification (OSV) by examining different loss functions for Convolutional Neural Network . Expand
Improving sensitivity of mercury detection using learning based smartphone colorimetry
In this work, a gadget was designed and printed using a lightweight 3D material, which can be attached to any smartphone and integrated with optical components. Expand
Assisted Excitation of Activations: A Learning Technique to Improve Object Detectors
We present a simple yet effective learning technique that significantly improves mAP of YOLO object detectors without compromising their speed. Expand