Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Nov 24, 2025

Decoding Natural Behavior from Neuroethological Embedding

Decoding Natural Behavior from Neuroethological Embedding

Published on: October 3, 2025

Toward Holistic Scene Understanding: Feedback Enabled Cascaded Classification Models.

Congcong Li¹, Adarsh Kowdle, Ashutosh Saxena

¹Department of Electrical and Computer Engineering, Cornell University, Ithaca, NY 14853-5401 USA. cl758@cornell.edu, tsuhan@ece.cornell.edu

IEEE Transactions on Pattern Analysis and Machine Intelligence

|December 7, 2011

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

A Nomogram for Predicting Successful Weaning from Invasive Mechanical Ventilation Withdrawal in Patients with Chronic Obstructive Pulmonary Disease Complicated by Respiratory Failure.

International journal of chronic obstructive pulmonary disease·2026

Same author

Assessing symptom clusters in post-total knee arthroplasty patients: development and validation of a new scale.

BMC musculoskeletal disorders·2026

Same author

Fused pyrazolo[3,4-<i>d</i>]pyridazine <i>N</i>-oxides with balanced detonation performance and safety.

Chemical communications (Cambridge, England)·2026

Same author

Symptom experience of patients after total knee arthroplasty in China: a longitudinal qualitative study.

BMJ open·2026

Same author

Impact of school physical activity on physical and mental health in children and adolescents: A Meta-analysis based on RE-AIM.

European child & adolescent psychiatry·2026

Same author

Small airway dysfunction: a shared, early, measurable, and potentially treatable trait across COPD, asthma, and fibrotic lung disease.

Frontiers in medicine·2026

Same journal

Relation DETR+: Exploring Explicit Position Relation Prior for Dense Prediction.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

RBF++: Quantifying and Optimizing Reasoning Boundaries across Measurable and Unmeasurable Capabilities for Chain-of-Thought Reasoning.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

CAFE: Cross-View Adaptive Fusion and Cluster Center Enhancement for Robust Multi-View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

Ethics-Aware Safe Reinforcement Learning for Rare-Event Risk Control in Interactive Urban Driving.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

Learning Shape Anchors for Holistic Indoor Scene Understanding.

IEEE transactions on pattern analysis and machine intelligence·2026

See all related articles

This study introduces Feedback Enabled Cascaded Classification Models (FE-CCM) to improve scene understanding tasks. The novel approach enhances performance across multiple computer vision subtasks by enabling classifiers to share information without altering their internal workings.

Area of Science:

Computer Vision
Machine Learning
Artificial Intelligence

Background:

Scene understanding involves complex subtasks like object detection and depth estimation.
Existing classifiers for these tasks are often state-of-the-art but operate independently.
There's a need for algorithms that leverage correlations between classifier outputs without modifying them.

Purpose of the Study:

To develop a novel algorithm, Feedback Enabled Cascaded Classification Models (FE-CCM), for joint optimization of scene understanding subtasks.
To capture correlations between classifier outputs using a black-box approach.
To improve performance in various scene understanding tasks and robotic applications.

Main Methods:

Proposed Feedback Enabled Cascaded Classification Models (FE-CCM).

More Related Videos

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Related Experiment Videos

Last Updated: Nov 24, 2025

Decoding Natural Behavior from Neuroethological Embedding

Decoding Natural Behavior from Neuroethological Embedding

Published on: October 3, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Employed a two-layer cascade of classifiers, feeding outputs from the first layer to the second.

Implemented a feedback mechanism during training for error mode focus.

Main Results:

Significantly improved performance across multiple scene understanding subtasks: depth estimation, scene categorization, event categorization, object detection, geometric labeling, and saliency detection.
Demonstrated enhanced performance in robotic applications, including object grasping and object finding.
Validated the effectiveness of the FE-CCM approach in leveraging inter-task correlations.

Conclusions:

FE-CCM effectively integrates multiple scene understanding classifiers.
The feedback mechanism enhances learning by directing attention to specific error modes.
The proposed method offers a versatile and powerful approach for improving computer vision and robotics tasks.