Papers › Visual Coreference Resolution in Visual Dialog using Neural Module Networks

Visual Coreference Resolution in Visual Dialog using Neural Module Networks

6 Sep 2018ECCV 2018 9arXiv:1809.01816archive 2025-07-28

Satwik Kottur, José M. F. Moura, Devi Parikh, Dhruv Batra, Marcus Rohrbach

Visual dialog entails answering a series of questions grounded in an image, using dialog history as context. In addition to the challenges found in visual question answering (VQA), which can be seen as one-round dialog, visual dialog encompasses several more. We focus on one such problem called visual coreference resolution that involves determining which words, typically noun phrases and pronouns, co-refer to the same entity/object instance in an image. This is crucial, especially for pronouns (e.g., `it'), as the dialog agent must first link it to a previous coreference (e.g., `boat'), and only then can rely on the visual grounding of the coreference `boat' to reason about the pronoun `it'. Prior work (in visual dialog) models visual coreference resolution either (a) implicitly via a memory network over history, or (b) at a coarse level for the entire question; and not explicitly at a phrase level of granularity. In this work, we propose a neural module network architecture for visual dialog by introducing two novel modules - Refer and Exclude - that perform explicit, grounded, coreference resolution at a finer word level. We demonstrate the effectiveness of our model on MNIST Dialog, a visually simple yet coreference-wise complex dataset, by achieving near perfect accuracy, and on VisDial, a large and challenging visual dialog dataset on real images, where our model outperforms other approaches, and is more interpretable, grounded, and consistent qualitatively.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

facebookresearch/corefnmn mentioned on GitHubtfNOASSERTION report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Common Sense ReasoningCoreference ResolutionVisual DialogVisual GroundingVisual Question AnsweringVisual Question Answering (VQA)coreference-resolution

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Common Sense Reasoning Visual Dialog v0.9 NMN [kottur2018visual] 1 in 10 R@5 80.1 #1 of 1 Archive leaderboard report
Visual Dialog VisDial v0.9 val CorefNMN (ResNet-152) MRR 64.1 #3 of 18 Archive leaderboard report
Visual Dialog VisDial v0.9 val CorefNMN (ResNet-152) Mean Rank 4.45 #3 of 18 Archive leaderboard report
Visual Dialog VisDial v0.9 val CorefNMN (ResNet-152) R@1 50.92 #3 of 18 Archive leaderboard report
Visual Dialog VisDial v0.9 val CorefNMN (ResNet-152) R@10 88.81 #3 of 18 Archive leaderboard report
Visual Dialog VisDial v0.9 val CorefNMN (ResNet-152) R@5 80.18 #3 of 18 Archive leaderboard report
Visual Dialog VisDial v0.9 val CorefNMN MRR 63.6 #5 of 18 Archive leaderboard report
Visual Dialog VisDial v0.9 val CorefNMN Mean Rank 4.53 #5 of 18 Archive leaderboard report
Visual Dialog VisDial v0.9 val CorefNMN R@1 50.24 #5 of 18 Archive leaderboard report
Visual Dialog VisDial v0.9 val CorefNMN R@10 88.51 #5 of 18 Archive leaderboard report
Visual Dialog VisDial v0.9 val CorefNMN R@5 79.81 #5 of 18 Archive leaderboard report
Visual Dialog Visual Dialog v1.0 test-std CorefNMN (ResNet-152) MRR (x 100) 61.50 #67 of 80 Archive leaderboard report
Visual Dialog Visual Dialog v1.0 test-std CorefNMN (ResNet-152) Mean 4.40 #67 of 80 Archive leaderboard report
Visual Dialog Visual Dialog v1.0 test-std CorefNMN (ResNet-152) NDCG (x 100) 54.70 #67 of 80 Archive leaderboard report
Visual Dialog Visual Dialog v1.0 test-std CorefNMN (ResNet-152) R@1 47.55 #67 of 80 Archive leaderboard report
Visual Dialog Visual Dialog v1.0 test-std CorefNMN (ResNet-152) R@10 88.80 #67 of 80 Archive leaderboard report
Visual Dialog Visual Dialog v1.0 test-std CorefNMN (ResNet-152) R@5 78.10 #67 of 80 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Memory Network

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections