mlsanityData to 7 Oct 2026

Multi-scale recognition with DAG-CNNs

arXiv 1505.05232 ↗First posted 20 May 2015cs.CVcited by 23

Not on the map. Why?

Abstract (arXiv)

We explore multi-scale convolutional neural nets (CNNs) for image classification. Contemporary approaches extract features from a single output layer. By extracting features from multiple layers, one can simultaneously reason about high, mid, and low-level features during classification. The resulting multi-scale architecture can itself be seen as a feed-forward model that is structured as a directed acyclic graph (DAG-CNNs). We use DAG-CNNs to learn a set of multiscale features that can be effectively shared between coarse and fine-grained classification tasks. While fine-tuning such models helps performance, we show that even "off-the-self" multiscale features perform quite well. We present extensive analysis and demonstrate state-of-the-art classification performance on three standard scene benchmarks (SUN397, MIT67, and Scene15). In terms of the heavily benchmarked MIT67 and Scene15 datasets, our results reduce the lowest previously-reported error by 23.9% and 9.5%, respectively.

Key facts. Cited by 23 papers in the map (data through 7 Oct 2026). First posted on arXiv 20 May 2015.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2021: 10; 2022: 7; 2023: 3; 2024: 3.

23 papers.

Sort:

  1. 2021CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classificationcited by 806
  2. 2021OpenGAN: Open-Set Recognition via Open Data Generationcited by 135
  3. 2022HiFormer: Hierarchical Multi-scale Representations Using Transformers for Medical Image Segmentationcited by 94
  4. 2023PCRLv2: A Unified Visual Information Preservation Framework for Self-supervised Pre-training in Medical Image Analysiscited by 40
  5. 2022Visual Attention Methods in Deep Learning: An In-Depth Surveycited by 37
  6. 2021Joint Visual Semantic Reasoning: Multi-Stage Decoder for Text Recognitioncited by 24
  7. 2021Deep Learning for Scene Classification: A Surveycited by 9
  8. 2021Probabilistic Graph Attention Network with Conditional Kernels for Pixel-Wise Predictioncited by 8
  9. 2021Multi-layered Semantic Representation Network for Multi-label Image Classificationcited by 6
  10. 2021MRC-LSTM: A Hybrid Approach of Multi-scale Residual CNN and LSTM to Predict Bitcoin Pricecited by 5
  11. 2023SRRM: Semantic Region Relation Model for Indoor Scene Recognitioncited by 5
  12. 2022ThreshNet: An Efficient DenseNet Using Threshold Mechanism to Reduce Connectionscited by 4
11 more
  1. 2024Leveraging Self-Supervised Learning for Scene Classification in Child Sexual Abuse Imagerycited by 4
  2. 2021Target Detection and Segmentation in Circular-Scan Synthetic-Aperture-Sonar Images using Semi-Supervised Convolutional Encoder-Decoderscited by 3
  3. 2021New Pruning Method Based on DenseNet Network for Image Classificationcited by 2
  4. 2022Recent Advances in Scene Image Representation and Classificationcited by 1
  5. 2023EnTri: Ensemble Learning with Tri-level Representations for Explainable Scene Recognitioncited by 1
  6. 2024Multi-task Prompt Words Learning for Social Media Content Generationcited by 1
  7. 2021Scene Retrieval for Contextual Visual Mappingcited by 0
  8. 2022Scene Recognition with Objectness, Attribute and Category Learningcited by 0
  9. 2022Look Closer to Your Enemy: Learning to Attack via Teacher-Student Mimickingcited by 0
  10. 2022Predicting Topological Maps for Visual Navigation in Unexplored Environmentscited by 0
  11. 2024Intuitionistic Fuzzy Cognitive Maps for Interpretable Image Classificationcited by 0