Papers › Attribution-aware Weight Transfer: A Warm-Start Initialization for Class-Incremental...

Attribution-aware Weight Transfer: A Warm-Start Initialization for Class-Incremental Semantic Segmentation

13 Oct 2022arXiv:2210.07207archive 2025-07-28

Dipam Goswami, René Schuster, Joost Van de Weijer, Didier Stricker

In class-incremental semantic segmentation (CISS), deep learning architectures suffer from the critical problems of catastrophic forgetting and semantic background shift. Although recent works focused on these issues, existing classifier initialization methods do not address the background shift problem and assign the same initialization weights to both background and new foreground class classifiers. We propose to address the background shift with a novel classifier initialization method which employs gradient-based attribution to identify the most relevant weights for new classes from the classifier's weights for the previous background and transfers these weights to the new classifier. This warm-start weight initialization provides a general solution applicable to several CISS methods. Furthermore, it accelerates learning of new classes while mitigating forgetting. Our experiments demonstrate significant improvement in mIoU compared to the state-of-the-art CISS methods on the Pascal-VOC 2012, ADE20K and Cityscapes datasets.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology: its repositories, every harvested function with whether it ran, its licence and the call to fetch it.

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2210.07207")

Code

Syntology Ran 6 of 8 code samples harvested from 1 repository linked to this paper; 2 have no recorded run. Of those that ran: 6 ran with no contract checked.

By repository: official repository: 8 samples from 1 repository, 6 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

dfki-av/awt-for-ciss officialmentioned in papermentioned on GitHubpytorchMIT 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

8 samples harvested; 6 ran; 0 honoured the contract we drafted; 2 have no recorded run. Read from Syntology's graph 2026-09-25; that is when this build read the record, not when the samples ran.

6ran
2unverified

Licence: 0 of the 8 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from dfki-av/awt-for-ciss. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

flip dfki-av/awt-for-ciss/segmentation_module.py official repository ran fingerprinted MIT (permissive) · 7f8043e7dd6c2735 · report
get_per_task_classes dfki-av/awt-for-ciss/tasks.py official repository ran MIT (permissive) · 178a3050bc16f2e5 · report
get_task_labels dfki-av/awt-for-ciss/tasks.py official repository ran MIT (permissive) · 8aaaf28fe82763c9 · report
get_task_labels_attr dfki-av/awt-for-ciss/tasks.py official repository ran MIT (permissive) · 0a93283e9a1abf38 · report
modify_command_options dfki-av/awt-for-ciss/argparser.py official repository ran MIT (permissive) · 8f85871d3af5c3c3 · report
soft_crossentropy dfki-av/awt-for-ciss/utils/loss.py official repository ran MIT (permissive) · 887fce52f90eb3b2 · report
get_loss dfki-av/awt-for-ciss/utils/loss.py official repository unverified MIT (permissive) · 90ca9d04a07dfe93 · report
nca dfki-av/awt-for-ciss/utils/loss.py official repository unverified MIT (permissive) · b494dd1e415df215 · report

Tasks

Class-Incremental Semantic SegmentationContinual LearningOverlapped 10-1Overlapped 100-10Overlapped 100-5Overlapped 100-50Overlapped 14-1Overlapped 15-1Overlapped 15-5Overlapped 5-3Overlapped 50-50Semantic Segmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Overlapped 10-1 Cityscapes MiB+AWT mIoU 44.9 #1 of 1 Archive leaderboard report
Overlapped 10-1 PASCAL VOC 2012 SSUL+AWT mIoU 60.7 #5 of 13 Archive leaderboard report
Overlapped 100-10 ADE20K MiB+AWT Mean IoU (test) 33.2 #3 of 6 Archive leaderboard report
Overlapped 100-5 ADE20K MiB+AWT mIoU 31.1 #4 of 8 Archive leaderboard report
Overlapped 100-50 ADE20K MiB+AWT mIoU 35.6 #2 of 7 Archive leaderboard report
Overlapped 14-1 Cityscapes MiB+AWT mIoU 46.9 #1 of 1 Archive leaderboard report
Overlapped 15-1 PASCAL VOC 2012 SSUL+AWT mIoU 67.6 #6 of 13 Archive leaderboard report
Overlapped 15-5 PASCAL VOC 2012 SSUL+AWT Mean IoU (val) 71.4 #6 of 13 Archive leaderboard report
Overlapped 5-3 PASCAL VOC 2012 SSUL+AWT Mean IoU (test) 57.1 #4 of 4 Archive leaderboard report
Overlapped 50-50 ADE20K MiB+AWT mIoU 33.5 #2 of 7 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.

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