REDLCV23
* *Resource Efficient Deep Learning for Computer Vision
* Accelerating Deep Neural Networks via Semi-Structured Activation Sparsity
* Accumulation Knowledge Distillation for Conditional GAN Compression
* AntiNODE: Evaluating Efficiency Robustness of Neural ODEs
* Bi-Encoder Cascades for Efficient Image Search
* Can Unstructured Pruning Reduce the Depth in Deep Neural Networks?
* Characterizing Face Recognition for Resource Efficient Deployment on Edge
* Comprehensive Study of Transfer Learning under Constraints, A
* CoroNetGAN: Controlled Pruning of GANs via Hypernetworks
* Cross-Domain Transfer Learning with CoRTe: Consistent and Reliable Transfer from Black-Box to Lightweight Segmentation Model
* Cross-model temporal cooperation via saliency maps for efficient frame classification
* Deterministic Neural Illumination Mapping for Efficient Auto-White Balance Correction
* DetOFA: Efficient Training of Once-for-All Networks for Object Detection using Path Filter
* Developing Robust and Lightweight Adversarial Defenders by Enforcing Orthogonality on Attack-Agnostic Denoising Autoencoders
* DONNAv2: Lightweight Neural Architecture Search for Vision tasks
* Dynamic Neural Network is All You Need: Understanding the Robustness of Dynamic Mechanisms in Neural Networks
* Efficient Neural PDE-Solvers using Quantization Aware Training
* Enhancing Differentiable Architecture Search: A Study on Small Number of Cell Blocks in the Search Stage, and Important Branches-based Cells Selection
* Entropic Score metric: Decoupling Topology and Size in Training-free NAS
* Extending TrOCR for Text Localization-Free OCR of Full-Page Scanned Receipt Images
* Factorized Dynamic Fully-Connected Layers for Neural Networks
* Fast Object Detection in High-Resolution Videos
* HyperSparse Neural Networks: Shifting Exploration to Exploitation through Adaptive Regularization
* InstaTune: Instantaneous Neural Architecture Search During Fine-Tuning
* Lightweight Vision Transformer with Spatial and Channel Enhanced Self-Attention
* MGiaD: Multigrid in all dimensions. Efficiency and robustness by weight sharing and coarsening in resolution and channel dimensions*
* MOFA: A Model Simplification Roadmap for Image Restoration on Mobile Devices
* Multi-Exit Resource-Efficient Neural Architecture for Image Classification with Optimized Fusion Block
* NCQS: Nonlinear Convex Quadrature Surrogate Hyperparameter Optimization
* Post Training Mixed Precision Quantization of Neural Networks using First-Order Information
* QBitOpt: Fast and Accurate Bitwidth Reallocation during Training
* Quantized Generative Models for Solving Inverse Problems
* Ray-Patch: An Efficient Querying for Light Field Transformers
* RCD-SGD: Resource-Constrained Distributed SGD in Heterogeneous Environment Via Submodular Partitioning
* RCV2023 Challenges: Benchmarking Model Training and Inference for Resource-Constrained Deep Learning
* Reconstructing Pruned Filters using Cheap Spatial Transformations
* Revisiting Kernel Temporal Segmentation as an Adaptive Tokenizer for Long-form Video Understanding
* SCoTTi: Save Computation at Training Time with an adaptive framework
* Shannon Strikes Again! Entropy-based Pruning in Deep Neural Networks for Transfer Learning under Extreme Memory and Computation Budgets
* Shapley Deep Learning: A Consensus for General-Purpose Vision Systems
* Simple and Generic Framework for Feature Distillation via Channel-wise Transformation, A
* Single-Shot Pruning for Pre-trained Models: Rethinking the Importance of Magnitude Pruning
* Softmax Bias Correction for Quantized Generative Models
* Surround the Nonlinearity: Inserting Foldable Convolutional Autoencoders to Reduce Activation Footprint
* Tiny and Efficient Model for the Edge Detection Generalization
* Video Action Recognition with Adaptive Zooming Using Motion Residuals
* When Layers Play the Lottery, all Tickets Win at Initialization
* YOLOBench: Benchmarking Efficient Object Detectors on Embedded Systems
* ZiCo-BC: A Bias Corrected Zero-Shot NAS for Vision Tasks
49 for REDLCV23