paper-with-me

Papers

Clockwork Convnets for Video Semantic Segmentation

2016-08-11 · Evan Shelhamer, Kate Rakelly, Judy Hoffman, Trevor Darrell

Recent years have seen tremendous progress in still-image segmentation; however the na\"ive application of these state-of-the-art algorithms to every video frame requires considerable computation and ignores the temporal continuity inherent in video. We propose a video recognition framework that relies on two key observations: 1) while pixels may change rapidly from frame to frame, the semantic content of a scene evolves more slowly, and 2) execution can be viewed as an aspect of architecture, yielding purpose-fit computation schedules for networks. We define a novel family of "clockwork" convnets driven by fixed or adaptive clock signals that schedule the processing of different layers at different update rates according to their semantic stability. We design a pipeline schedule to reduce latency for real-time recognition and a fixed-rate schedule to reduce overall computation. Finally, we extend clockwork scheduling to adaptive video processing by incorporating data-driven clocks that can be tuned on unlabeled video. The accuracy and efficiency of clockwork convnets are evaluated on the Youtube-Objects, NYUD, and Cityscapes video datasets.

📄 PDF Abstract BibTeX arXiv:1608.03609

Code (1)

shelhamer/clockwork-fcn caffe2

Tasks

Image SegmentationSchedulingSegmentationSemantic SegmentationVideo RecognitionVideo Semantic Segmentation

Similar Papers 제목 키워드 기반

Clockwork Variational Autoencoders

2021-02-18 · NeurIPS 2021 12 · Vaibhav Saxena, Jimmy Ba, Danijar Hafner

Deep learning has enabled algorithms to generate realistic images. However, accurately predicting long video sequences requires understanding long-term dependencies and remains an open challenge. While existing video pre…

MinecraftPredictionVideo Prediction

On the Importance of Label Quality for Semantic Segmentation

2018-06-01 · CVPR 2018 6 · Aleksandar Zlateski, Ronnachai Jaroensri, Prafull Sharma, Frédo Durand

Convolutional networks (ConvNets) have become the dominant approach to semantic image segmentation. Producing accurate, pixel--level labels required for this task is a tedious and time consuming process; however, produci…

Image SegmentationSemantic Segmentationvalid

Clockwork Diffusion: Efficient Generation With Model-Step Distillation

2023-12-13 · CVPR 2024 1 · Amirhossein Habibian, Amir Ghodrati, Noor Fathima, Guillaume Sautiere 외

This work aims to improve the efficiency of text-to-image diffusion models. While diffusion models use computationally expensive UNet-based denoising operations in every generation step, we identify that not all operatio…

DenoisingImage GenerationmodelText to Image Generation+1

Benchmarking Generative Latent Variable Models for Speech

2022-02-22 · Jakob D. Havtorn, Lasse Borgholt, Søren Hauberg, Jes Frellsen 외

Stochastic latent variable models (LVMs) achieve state-of-the-art performance on natural image generation but are still inferior to deterministic models on speech. In this paper, we develop a speech benchmark of popular …

BenchmarkingImage GenerationPhoneme RecognitionVideo Generation

MogaNet: Multi-order Gated Aggregation Network

2022-11-07 · Siyuan Li, Zedong Wang, Zicheng Liu, Cheng Tan 외

By contextualizing the kernel as global as possible, Modern ConvNets have shown great potential in computer vision tasks. However, recent progress on \textit{multi-order game-theoretic interaction} within deep neural net…

3D Human Pose EstimationImage ClassificationInstance Segmentationobject-detection+5