Picture for Maxim Berman

Maxim Berman

Revisiting Evaluation Metrics for Semantic Segmentation: Optimization and Evaluation of Fine-grained Intersection over Union

Add code
Oct 30, 2023
Viaarxiv icon

Spatial Consistency Loss for Training Multi-Label Classifiers from Single-Label Annotations

Add code
Mar 11, 2022
Figure 1 for Spatial Consistency Loss for Training Multi-Label Classifiers from Single-Label Annotations
Figure 2 for Spatial Consistency Loss for Training Multi-Label Classifiers from Single-Label Annotations
Figure 3 for Spatial Consistency Loss for Training Multi-Label Classifiers from Single-Label Annotations
Figure 4 for Spatial Consistency Loss for Training Multi-Label Classifiers from Single-Label Annotations
Viaarxiv icon

Optimization for Medical Image Segmentation: Theory and Practice when evaluating with Dice Score or Jaccard Index

Add code
Oct 26, 2020
Figure 1 for Optimization for Medical Image Segmentation: Theory and Practice when evaluating with Dice Score or Jaccard Index
Figure 2 for Optimization for Medical Image Segmentation: Theory and Practice when evaluating with Dice Score or Jaccard Index
Figure 3 for Optimization for Medical Image Segmentation: Theory and Practice when evaluating with Dice Score or Jaccard Index
Figure 4 for Optimization for Medical Image Segmentation: Theory and Practice when evaluating with Dice Score or Jaccard Index
Viaarxiv icon

AOWS: Adaptive and optimal network width search with latency constraints

Add code
May 21, 2020
Figure 1 for AOWS: Adaptive and optimal network width search with latency constraints
Figure 2 for AOWS: Adaptive and optimal network width search with latency constraints
Figure 3 for AOWS: Adaptive and optimal network width search with latency constraints
Figure 4 for AOWS: Adaptive and optimal network width search with latency constraints
Viaarxiv icon

Discriminative training of conditional random fields with probably submodular constraints

Add code
Nov 25, 2019
Figure 1 for Discriminative training of conditional random fields with probably submodular constraints
Figure 2 for Discriminative training of conditional random fields with probably submodular constraints
Figure 3 for Discriminative training of conditional random fields with probably submodular constraints
Figure 4 for Discriminative training of conditional random fields with probably submodular constraints
Viaarxiv icon

Optimizing the Dice Score and Jaccard Index for Medical Image Segmentation: Theory & Practice

Add code
Nov 05, 2019
Figure 1 for Optimizing the Dice Score and Jaccard Index for Medical Image Segmentation: Theory & Practice
Figure 2 for Optimizing the Dice Score and Jaccard Index for Medical Image Segmentation: Theory & Practice
Viaarxiv icon

Adaptive Compression-based Lifelong Learning

Add code
Jul 23, 2019
Figure 1 for Adaptive Compression-based Lifelong Learning
Figure 2 for Adaptive Compression-based Lifelong Learning
Figure 3 for Adaptive Compression-based Lifelong Learning
Figure 4 for Adaptive Compression-based Lifelong Learning
Viaarxiv icon

MultiGrain: a unified image embedding for classes and instances

Add code
Apr 03, 2019
Figure 1 for MultiGrain: a unified image embedding for classes and instances
Figure 2 for MultiGrain: a unified image embedding for classes and instances
Figure 3 for MultiGrain: a unified image embedding for classes and instances
Figure 4 for MultiGrain: a unified image embedding for classes and instances
Viaarxiv icon

Generating superpixels using deep image representations

Add code
Mar 11, 2019
Figure 1 for Generating superpixels using deep image representations
Figure 2 for Generating superpixels using deep image representations
Figure 3 for Generating superpixels using deep image representations
Figure 4 for Generating superpixels using deep image representations
Viaarxiv icon

Yes, IoU loss is submodular - as a function of the mispredictions

Add code
Sep 06, 2018
Viaarxiv icon